Class RiskMLInterface

java.lang.Object

neqsim.process.safety.risk.ml.RiskMLInterface

All Implemented Interfaces:

Serializable

public class RiskMLInterface
extends Object
implements Serializable

Machine Learning Integration Interface for Risk Assessment.

Provides a standardized interface for integrating external machine learning models with the NeqSim risk framework. Supports various ML use cases including:

• Failure prediction models
• Anomaly detection
• Remaining useful life (RUL) prediction
• Process optimization under risk constraints

Version:

1.0

Author:

NeqSim Development Team

See Also:

• Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static interface	RiskMLInterface.FeatureExtractor	Feature extractor for process data.
static class	RiskMLInterface.MLModel	Machine learning model wrapper.
static class	RiskMLInterface.MLPrediction	ML prediction result.
static interface	RiskMLInterface.MLPredictor	Functional interface for ML model prediction.
static class	RiskMLInterface.ModelPerformanceMetrics	Model performance metrics.
static class	RiskMLInterface.PredictionRecord	Prediction record for history tracking.

Field Summary

Fields

Modifier and Type	Field	Description
private Map<String, RiskMLInterface.FeatureExtractor>	featureExtractors	Feature extractors.
private int	maxHistorySize	Maximum history size.
private Map<String, RiskMLInterface.MLModel>	models	Registered ML models.
private String	name	Interface name.
private List<RiskMLInterface.PredictionRecord>	predictionHistory	Prediction history.
private static final long	serialVersionUID

Constructor Summary

Constructors

Constructor	Description
RiskMLInterface(String name)	Creates an ML interface.

Method Summary

Modifier and Type	Method	Description
RiskMLInterface.MLModel	createAnomalyDetectionModel(String modelId, String modelName)	Creates and registers an anomaly detection model.
RiskMLInterface.MLModel	createFailurePredictionModel(String modelId, String modelName)	Creates and registers a failure prediction model.
RiskMLInterface.MLModel	createRULModel(String modelId, String modelName)	Creates and registers an RUL prediction model.
List<RiskMLInterface.MLModel>	getActiveModels()
RiskMLInterface.MLModel	getModel(String modelId)
RiskMLInterface.ModelPerformanceMetrics	getModelPerformance(String modelId)	Gets model performance metrics.
List<RiskMLInterface.MLModel>	getModels()
String	getName()
RiskMLInterface.MLPrediction	predict(String modelId, Map<String,Double> features)	Makes a prediction using a registered model.
RiskMLInterface.MLPrediction	predictWithExtraction(String modelId, String extractorName, Map<String,Object> processData)	Makes a prediction with feature extraction.
void	provideFeedback(Instant predictionTimestamp, double actualValue)	Provides feedback on a prediction (for model improvement).
void	registerFeatureExtractor(String name, RiskMLInterface.FeatureExtractor extractor)	Registers a feature extractor.
void	registerModel(RiskMLInterface.MLModel model)	Registers an ML model.
String	toJson()	Converts to JSON string.
Map<String,Object>	toMap()	Converts to map for JSON serialization.
String	toString()

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Details

serialVersionUID

private static final long serialVersionUID

See Also:

• Constant Field Values

name

private String name

Interface name.

models

private Map<String, RiskMLInterface.MLModel> models

Registered ML models.

featureExtractors

private Map<String, RiskMLInterface.FeatureExtractor> featureExtractors

Feature extractors.

predictionHistory

private List<RiskMLInterface.PredictionRecord> predictionHistory

Prediction history.

maxHistorySize

private int maxHistorySize

Maximum history size.

Constructor Details

RiskMLInterface

public RiskMLInterface(String name)

Creates an ML interface.

Parameters:

name - interface name

Method Details

registerModel

public void registerModel(RiskMLInterface.MLModel model)

Registers an ML model.

Parameters:

model - ML model

createFailurePredictionModel

public RiskMLInterface.MLModel createFailurePredictionModel(String modelId,
 String modelName)

Creates and registers a failure prediction model.

Parameters:

modelId - model ID

modelName - model name

Returns:

created model

createAnomalyDetectionModel

public RiskMLInterface.MLModel createAnomalyDetectionModel(String modelId,
 String modelName)

Creates and registers an anomaly detection model.

Parameters:

modelId - model ID

modelName - model name

Returns:

created model

createRULModel

public RiskMLInterface.MLModel createRULModel(String modelId,
 String modelName)

Creates and registers an RUL prediction model.

Parameters:

modelId - model ID

modelName - model name

Returns:

created model

registerFeatureExtractor

public void registerFeatureExtractor(String name,
 RiskMLInterface.FeatureExtractor extractor)

Registers a feature extractor.

Parameters:

name - extractor name

extractor - feature extractor

predict

public RiskMLInterface.MLPrediction predict(String modelId,
 Map<String,Double> features)

Makes a prediction using a registered model.

Parameters:

modelId - model ID

features - input features

Returns:

prediction result

predictWithExtraction

public RiskMLInterface.MLPrediction predictWithExtraction(String modelId,
 String extractorName,
 Map<String,Object> processData)

Makes a prediction with feature extraction.

Parameters:

modelId - model ID

extractorName - feature extractor name

processData - raw process data

Returns:

prediction result

provideFeedback

public void provideFeedback(Instant predictionTimestamp,
 double actualValue)

Provides feedback on a prediction (for model improvement).

Parameters:

predictionTimestamp - timestamp of prediction

actualValue - actual observed value

getModelPerformance

public RiskMLInterface.ModelPerformanceMetrics getModelPerformance(String modelId)

Gets model performance metrics.

Parameters:

modelId - model ID

Returns:

performance metrics

getName

public String getName()

getModel

public RiskMLInterface.MLModel getModel(String modelId)

getModels

public List<RiskMLInterface.MLModel> getModels()

getActiveModels

public List<RiskMLInterface.MLModel> getActiveModels()

toMap

public Map<String,Object> toMap()

Converts to map for JSON serialization.

Returns:

map representation

toJson

public String toJson()

Converts to JSON string.

Returns:

JSON representation

toString

public String toString()

Overrides:

toString in class Object