MCP 指南deprecated

MCP 历史工具设计（未公开）

保留的 MCP 工具设计草案，不代表 RailWise 已对外提供任何工具、参数或返回数据。

复核 2026-07-20入门不进入公开索引RailWise 技术团队

MCP 指南

RailWise MCP Server 工具列表

1. 工具概述

RailWise MCP Server 目前提供 12 个工具，分为四大类别：

类别	工具数量	说明
数据查询	4 个	监测数据、项目信息、设备状态、历史记录查询
计算分析	3 个	坐标转换、平差计算、变形分析
项目管理	3 个	项目查询、测点管理、任务调度
系统管理	2 个	用户权限、系统状态查询

所有工具均遵循 MCP 协议规范，使用 JSON Schema 定义参数和返回值结构。

2. 数据查询工具

2.1 `railwise_query_monitoring_data` — 查询监测数据

查询全站仪自动化监测系统的实时或历史监测数据。

参数定义：

{
  "type": "object",
  "properties": {
    "project_id": {
      "type": "string",
      "description": "项目唯一标识符",
      "pattern": "^PRJ-[A-Z0-9]{8}$"
    },
    "point_ids": {
      "type": "array",
      "items": { "type": "string" },
      "description": "测点ID列表，为空则查询所有测点"
    },
    "data_type": {
      "type": "string",
      "enum": ["settlement", "displacement", "convergence", "stress", "all"],
      "description": "数据类型：沉降、位移、收敛、应力、全部"
    },
    "start_time": {
      "type": "string",
      "format": "date-time",
      "description": "查询起始时间，ISO 8601 格式"
    },
    "end_time": {
      "type": "string",
      "format": "date-time",
      "description": "查询结束时间，ISO 8601 格式"
    },
    "aggregation": {
      "type": "string",
      "enum": ["raw", "hourly", "daily", "weekly"],
      "description": "数据聚合方式",
      "default": "raw"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 10000,
      "description": "返回记录数量上限",
      "default": 1000
    }
  },
  "required": ["project_id", "start_time", "end_time"]
}

返回值结构：

{
  "project_id": "PRJ-A1B2C3D4",
  "data_type": "settlement",
  "time_range": {
    "start": "2025-01-10T00:00:00Z",
    "end": "2025-01-15T23:59:59Z"
  },
  "total_records": 1248,
  "points": [
    {
      "point_id": "PT-001",
      "point_name": "K3+520-左线-1",
      "point_type": "settlement",
      "records": [
        {
          "timestamp": "2025-01-15T08:00:00Z",
          "value": -12.3,
          "unit": "mm",
          "change_rate": -0.8,
          "warning_level": "orange",
          "status": "normal"
        }
      ]
    }
  ],
  "summary": {
    "normal_count": 1240,
    "yellow_warning_count": 6,
    "orange_warning_count": 2,
    "red_warning_count": 0
  }
}

使用示例：

// AI 助手调用示例
{
  "name": "railwise_query_monitoring_data",
  "arguments": {
    "project_id": "PRJ-A1B2C3D4",
    "data_type": "settlement",
    "start_time": "2025-01-14T00:00:00Z",
    "end_time": "2025-01-15T23:59:59Z",
    "aggregation": "hourly",
    "limit": 500
  }
}

2.2 `railwise_query_project_info` — 查询项目信息

查询监测项目的基本信息、测点配置和监测方案。

参数定义：

{
  "type": "object",
  "properties": {
    "project_id": {
      "type": "string",
      "description": "项目唯一标识符"
    },
    "include_points": {
      "type": "boolean",
      "description": "是否包含测点详情",
      "default": true
    },
    "include_devices": {
      "type": "boolean",
      "description": "是否包含设备信息",
      "default": false
    }
  },
  "required": ["project_id"]
}

返回值结构：

{
  "project_id": "PRJ-A1B2C3D4",
  "project_name": "宁波地铁3号线控制保护区监测",
  "project_type": "subway_protection",
  "status": "active",
  "location": {
    "province": "浙江省",
    "city": "宁波市",
    "district": "鄞州区"
  },
  "client": "宁波市轨道交通集团",
  "contractor": "中铁一局",
  "monitor_unit": "宁波睿威工程技术有限公司",
  "start_date": "2024-06-01",
  "expected_end_date": "2025-12-31",
  "points": [
    {
      "point_id": "PT-001",
      "point_name": "K3+520-左线-1",
      "point_type": "settlement",
      "coordinate": {
        "x": 384567.123,
        "y": 3156784.456,
        "z": 12.345
      },
      "warning_threshold": {
        "yellow": -10.0,
        "orange": -15.0,
        "red": -20.0
      }
    }
  ],
  "monitoring_scheme": {
    "frequency": "4h",
    "instrument": "total_station",
    "data_retention_days": 365
  }
}

2.3 `railwise_query_device_status` — 查询设备状态

查询全站仪等监测设备的实时状态和运行信息。

参数定义：

{
  "type": "object",
  "properties": {
    "project_id": {
      "type": "string",
      "description": "项目ID，为空则查询所有可访问设备"
    },
    "device_ids": {
      "type": "array",
      "items": { "type": "string" },
      "description": "设备ID列表"
    }
  }
}

返回值结构：

{
  "devices": [
    {
      "device_id": "DEV-TS-001",
      "device_name": "全站仪-01",
      "device_type": "total_station",
      "model": "Leica TS16",
      "status": "online",
      "last_online": "2025-01-15T14:30:00Z",
      "project_id": "PRJ-A1B2C3D4",
      "location": "K3+520控制点",
      "battery_level": 78,
      "temperature": 15.2,
      "last_measurement": "2025-01-15T14:00:00Z",
      "measurement_count_today": 36,
      "firmware_version": "2.5.1"
    }
  ],
  "summary": {
    "total": 5,
    "online": 4,
    "offline": 1,
    "warning": 0
  }
}

2.4 `railwise_query_measurement_history` — 查询测量历史

查询特定测点的详细测量历史记录，包括原始观测数据。

参数定义：

{
  "type": "object",
  "properties": {
    "point_id": {
      "type": "string",
      "description": "测点ID"
    },
    "start_time": {
      "type": "string",
      "format": "date-time"
    },
    "end_time": {
      "type": "string",
      "format": "date-time"
    },
    "include_raw": {
      "type": "boolean",
      "description": "是否包含原始观测数据",
      "default": false
    }
  },
  "required": ["point_id", "start_time", "end_time"]
}

3. 计算分析工具

3.1 `railwise_coordinate_transform` — 坐标转换

执行不同坐标系之间的坐标转换，支持常见的工程测量坐标转换需求。

参数定义：

{
  "type": "object",
  "properties": {
    "coordinates": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "x": { "type": "number" },
          "y": { "type": "number" },
          "z": { "type": "number" }
        },
        "required": ["x", "y"]
      },
      "description": "待转换的坐标列表"
    },
    "source_crs": {
      "type": "string",
      "description": "源坐标系代码",
      "enum": ["CGCS2000", "WGS84", "Beijing54", "Xian80", "local"]
    },
    "target_crs": {
      "type": "string",
      "description": "目标坐标系代码",
      "enum": ["CGCS2000", "WGS84", "Beijing54", "Xian80", "local"]
    },
    "transform_method": {
      "type": "string",
      "description": "转换方法",
      "enum": ["three_parameter", "seven_parameter", "grid", "custom"],
      "default": "seven_parameter"
    },
    "transform_params": {
      "type": "object",
      "description": "转换参数（当使用自定义转换时）"
    }
  },
  "required": ["coordinates", "source_crs", "target_crs"]
}

返回值结构：

{
  "transform_id": "TRF-20250115-001",
  "source_crs": "CGCS2000",
  "target_crs": "local",
  "transform_method": "seven_parameter",
  "results": [
    {
      "source": { "x": 384567.123, "y": 3156784.456, "z": 12.345 },
      "target": { "x": 1000.000, "y": 2000.000, "z": 12.345 },
      "residual": { "dx": 0.002, "dy": 0.001, "dz": 0.000 }
    }
  ],
  "statistics": {
    "point_count": 1,
    "max_residual": 0.002,
    "rms": 0.0015
  }
}

3.2 `railwise_adjustment_compute` — 平差计算

执行测量平差计算，支持条件平差、间接平差等多种平差模型。

参数定义：

{
  "type": "object",
  "properties": {
    "adjustment_type": {
      "type": "string",
      "enum": ["indirect", "conditional", "indirect_with_constraints"],
      "description": "平差类型"
    },
    "observations": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "obs_id": { "type": "string" },
          "obs_type": { "type": "string", "enum": ["distance", "angle", "elevation", "coordinate"] },
          "value": { "type": "number" },
          "precision": { "type": "number", "description": "观测精度（中误差）" }
        }
      }
    },
    "unknowns": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "unknown_id": { "type": "string" },
          "unknown_type": { "type": "string", "enum": ["coordinate", "orientation", "parameter"] },
          "approximate_value": { "type": "number" }
        }
      }
    }
  },
  "required": ["adjustment_type", "observations", "unknowns"]
}

3.3 `railwise_deformation_analysis` — 变形分析

对监测数据进行变形分析，计算累计变形量、变化速率、趋势预测等。

参数定义：

{
  "type": "object",
  "properties": {
    "point_id": {
      "type": "string",
      "description": "测点ID"
    },
    "analysis_type": {
      "type": "string",
      "enum": ["cumulative", "rate", "trend", "forecast", "comprehensive"],
      "description": "分析类型"
    },
    "start_time": {
      "type": "string",
      "format": "date-time"
    },
    "end_time": {
      "type": "string",
      "format": "date-time"
    },
    "forecast_days": {
      "type": "integer",
      "description": "预测天数（当 analysis_type 为 forecast 时）",
      "default": 7
    }
  },
  "required": ["point_id", "analysis_type", "start_time", "end_time"]
}

返回值结构（综合分析）：

{
  "point_id": "PT-001",
  "analysis_type": "comprehensive",
  "time_range": {
    "start": "2024-06-01T00:00:00Z",
    "end": "2025-01-15T23:59:59Z"
  },
  "cumulative_deformation": {
    "max_value": -12.3,
    "min_value": -0.5,
    "current_value": -12.3,
    "unit": "mm"
  },
  "deformation_rate": {
    "current_rate": -0.8,
    "average_rate": -0.5,
    "max_rate": -1.2,
    "unit": "mm/day"
  },
  "trend": {
    "direction": "settling",
    "stability": "unstable",
    "confidence": 0.85
  },
  "forecast": {
    "forecast_days": 7,
    "predicted_value": -14.1,
    "confidence_interval": {
      "lower": -15.5,
      "upper": -12.7
    }
  },
  "warning_assessment": {
    "current_level": "orange",
    "trend_level": "red",
    "recommendation": "建议加密监测频率至2小时/次，并启动现场巡查"
  }
}

4. 项目管理工具

4.1 `railwise_list_projects` — 列出项目

列出当前用户有权限访问的监测项目列表。

参数定义：

{
  "type": "object",
  "properties": {
    "status": {
      "type": "string",
      "enum": ["active", "paused", "completed", "all"],
      "default": "active"
    },
    "limit": {
      "type": "integer",
      "default": 50,
      "maximum": 200
    }
  }
}

4.2 `railwise_query_point_details` — 查询测点详情

查询特定测点的详细配置信息和历史状态。

参数定义：

{
  "type": "object",
  "properties": {
    "point_id": {
      "type": "string",
      "description": "测点ID"
    }
  },
  "required": ["point_id"]
}

4.3 `railwise_query_task_schedule` — 查询任务调度

查询监测任务的调度计划和执行状态。

参数定义：

{
  "type": "object",
  "properties": {
    "project_id": {
      "type": "string"
    },
    "date": {
      "type": "string",
      "format": "date",
      "description": "查询日期，默认为今天"
    }
  },
  "required": ["project_id"]
}

5. 系统管理工具

5.1 `railwise_query_user_permissions` — 查询用户权限

查询当前 API Key 的权限范围和访问限制。

参数定义：

{
  "type": "object",
  "properties": {}
}

返回值结构：

{
  "api_key_id": "KEY-xxxxxxxx",
  "key_name": "Claude Desktop 集成",
  "permissions": {
    "data_read": true,
    "data_write": false,
    "project_read": true,
    "project_write": false,
    "device_read": true,
    "device_control": false,
    "compute": true,
    "admin": false
  },
  "accessible_projects": ["PRJ-A1B2C3D4", "PRJ-E5F6G7H8"],
  "rate_limit": {
    "requests_per_minute": 60,
    "requests_per_hour": 1000,
    "requests_per_day": 10000
  },
  "expires_at": "2025-12-31T23:59:59Z"
}

5.2 `railwise_query_system_status` — 查询系统状态

查询 RailWise 系统的整体运行状态和健康信息。

参数定义：

{
  "type": "object",
  "properties": {}
}

返回值结构：

{
  "system_status": "healthy",
  "version": "2.5.1",
  "services": {
    "api_gateway": { "status": "online", "latency_ms": 45 },
    "data_service": { "status": "online", "latency_ms": 120 },
    "compute_service": { "status": "online", "latency_ms": 80 },
    "device_service": { "status": "online", "latency_ms": 200 }
  },
  "statistics": {
    "active_projects": 156,
    "online_devices": 423,
    "today_measurements": 125840
  }
}

6. 工具调用完整示例

以下是一个典型的 AI 助手与 RailWise MCP Server 的交互流程：

6.1 用户请求

用户：请分析宁波地铁3号线 K3+520 断面左线1号测点最近一周的沉降趋势，并预测未来3天的变化。

6.2 AI 助手工具调用序列

Step 1: 查询项目信息（确认项目ID）
┌────────────────────────────────────────┐
│ Tool: railwise_list_projects           │
│ Arguments: { "status": "active" }        │
└────────────────────────────────────────┘

Step 2: 查询测点详情（确认测点ID）
┌────────────────────────────────────────┐
│ Tool: railwise_query_project_info      │
│ Arguments: {                           │
│   "project_id": "PRJ-A1B2C3D4",        │
│   "include_points": true               │
│ }                                      │
└────────────────────────────────────────┘

Step 3: 执行变形分析
┌────────────────────────────────────────┐
│ Tool: railwise_deformation_analysis      │
│ Arguments: {                           │
│   "point_id": "PT-001",                │
│   "analysis_type": "comprehensive",    │
│   "start_time": "2025-01-08T00:00:00Z",│
│   "end_time": "2025-01-15T23:59:59Z",  │
│   "forecast_days": 3                   │
│ }                                      │
└────────────────────────────────────────┘

6.3 AI 助手回复

AI：根据分析结果，宁波地铁3号线 K3+520 断面左线1号测点（PT-001）最近一周的沉降情况如下：

累计沉降：-12.3mm（当前值） 变化速率：-0.8mm/天（当前），平均 -0.5mm/天 趋势判断：沉降趋势不稳定，速率有加快迹象

未来3天预测：

预测沉降值：-14.1mm

置信区间：-15.5mm ~ -12.7mm

预警评估：

当前预警级别：橙色预警

趋势预警级别：红色预警

建议：加密监测频率至2小时/次，启动现场巡查，关注周边施工活动

7. 相关文档

MCP Server 概述 — 功能介绍和适用场景
MCP Server 安装配置 — 详细的安装步骤
MCP Server 资源访问 — 可访问的资源类型和路径格式
MCP Server 集成示例 — 与主流 AI 平台的集成教程
MCP Server 安全与认证 — 认证方式和权限控制

本文档由 RailWise 技术文档团队维护，最后更新于 2025-01-15。

MCP 历史工具设计（未公开）

RailWise MCP Server 工具列表

1. 工具概述

2. 数据查询工具

2.1 railwise_query_monitoring_data — 查询监测数据

2.2 railwise_query_project_info — 查询项目信息

2.3 railwise_query_device_status — 查询设备状态

2.4 railwise_query_measurement_history — 查询测量历史

3. 计算分析工具

3.1 railwise_coordinate_transform — 坐标转换

3.2 railwise_adjustment_compute — 平差计算

3.3 railwise_deformation_analysis — 变形分析

4. 项目管理工具

4.1 railwise_list_projects — 列出项目

4.2 railwise_query_point_details — 查询测点详情

4.3 railwise_query_task_schedule — 查询任务调度

5. 系统管理工具

5.1 railwise_query_user_permissions — 查询用户权限

5.2 railwise_query_system_status — 查询系统状态

6. 工具调用完整示例

6.1 用户请求

6.2 AI 助手工具调用序列

6.3 AI 助手回复

7. 相关文档

2.1 `railwise_query_monitoring_data` — 查询监测数据

2.2 `railwise_query_project_info` — 查询项目信息

2.3 `railwise_query_device_status` — 查询设备状态

2.4 `railwise_query_measurement_history` — 查询测量历史

3.1 `railwise_coordinate_transform` — 坐标转换

3.2 `railwise_adjustment_compute` — 平差计算

3.3 `railwise_deformation_analysis` — 变形分析

4.1 `railwise_list_projects` — 列出项目

4.2 `railwise_query_point_details` — 查询测点详情

4.3 `railwise_query_task_schedule` — 查询任务调度

5.1 `railwise_query_user_permissions` — 查询用户权限

5.2 `railwise_query_system_status` — 查询系统状态