Deploying machine learning models at cognitive devices for resources management for enterprises

Abstract

A platform and method for using cognition for managing enterprise energy needs. Cognitive platform allows for dynamic management of energy consumption, demand and baseline calculations through a cognitive platform and cognitive device. Employees as well as internal and external stakeholders can set performance indicators and monitor the parameters against the energy performance indicators. Based on the initial knowledge, the system identifies improvements in order to reach the energy key performance indicators. Depending on the feedback, the system learns and improves the accuracy of the predictions and suits them to a given industry or given enterprise scenario. Enterprise-wide energy or environmental management covers policies, planning, key performance indicators, goals, targets, works flows, user management, asset mapping, input-output energy flows, conservation options, performance management, analytics. The system and method allow for monitoring and verification by internal or external stakeholders.

We claim: 
     
         1 . A cognitive device comprising:
 a communication module configured to exchange instructions and configuration information related to at least one machine learning model with an external system;   a sensor configured to sense at least one input parameter related to a resource of an equipment or an equipment controller;   an adaptor configured to process information related to the equipment;   a processor configured to determine at least one action based on at least one learning operation utilizing at least a portion of the machine learning model and sensed input related to the equipment or the equipment controller;   an actuator configured to communicate the determined action to the equipment or the equipment controller.   
     
     
         2 . The cognitive device of  claim 1 , wherein the cognitive device is configured to be modified by the instructions and configuration information received from the external system to act as a master or the slave device to another cognitive device. 
     
     
         3 . The cognitive device of  claim 1 , wherein the instructions and configuration information exchanged with external system are related to version information, version selection, version updates, version upgrades, bug fixes, or model modifications related to the cognitive device. 
     
     
         4 . The cognitive device of  claim 1 , wherein the behavior of the hardware of cognitive device can be modified by the instructions and configuration information received from external system. 
     
     
         5 . The cognitive device of  claim 1 , wherein sensed input parameter is related to a furnace, air conditioner, heater, motor, fan, blower, pump, compressor, boiler, meter, heat exchanger, computer, computer software, router, antenna, switch, communication software, transmission device, smart meter, wire-line network or a wireless network. 
     
     
         6 . The cognitive device of  claim 1 , wherein the determined action addresses any one or more optimizations of energy consumption, maintenance, load, resource demand, resource supply, or resource usage. 
     
     
         7 . The cognitive device of  claim 1 , wherein the determined action addresses any one or more resource usages selected from: power off, power on, idle state, sleep state, CPU utilization, memory utilization, spectrum allocation or bandwidth allocation. 
     
     
         8 . A cognitive platform comprising:
 a service control module configured to exchange at least one set of instructions and configuration information related to a machine learning model with an external cognitive device;   a data receiving and data transmitting module configured to exchange data with the external cognitive device;   a data translation and interpretation module configured to translate and interpret instructions, configuration information and data exchanged between the external cognitive device and the cognitive platform;   a cognitive decision maker module configured to select at least one machine learning model from a plurality of machine learning models available and exchange instructions, configuration information and data related to deploying the selected machine learning model to the external cognitive device.   
     
     
         9 . The cognitive platform of  claim 8 , wherein the cognitive decision maker performing modifications to the selected machine learning model to optimize the performance of the machine learning model based on the hardware platform of the external cognitive device. 
     
     
         10 . The cognitive platform of  claim 8 , further comprising:
 an asset management module configured to exchange instructions and configuration information related to at least one sensor or one actuator to modify the behavior of the external cognitive device.   
     
     
         11 . The cognitive platform of  claim 8 , wherein the cognitive platform is configured to act as a master or a slave to another cognitive platform. 
     
     
         12 . The cognitive platform of  claim 9 , wherein the instructions and configuration information exchanged with external cognitive device are related to version information, version selection, version updates, version upgrades, bug fixes, machine learning model modifications, sensors or actuators. 
     
     
         13 . A method to deploying machine learning models to a cognitive device, the method comprising:
 exchanging instructions and configuration information related to at least one machine learning model with an external system;   obtaining at least one input parameter related to a resource of an equipment or an equipment controller via a sensor;   obtaining information related to the equipment;   determining at least one action based on at least one learning operation utilizing at least a portion of the machine learning model and sensed input related to the equipment or the equipment controller via a processor;   communicating the determined action to the equipment or the equipment controller via an actuator.   
     
     
         14 . The method of  claim 13 , wherein the cognitive device is modified by the instructions and configuration information received from external system to act as a master or a slave device to another cognitive device. 
     
     
         15 . The method of  claim 13 , wherein the instructions and configuration information exchanged with external system are related to version information, version selection, version updates, version upgrades, bug fixes, or model modifications related to the cognitive device. 
     
     
         16 . The method of  claim 13 , wherein the behavior of the hardware of cognitive device can be modified by the instructions and configuration information received from external system. 
     
     
         17 . The method of  claim 13 , wherein sensed input parameter is related to a furnace, air conditioner, heater, motor, fan, blower, pump, compressor, boiler, meter, heat exchanger, computer, computer software, router, antenna, switch, communication software, transmission device, smart meter, wire-line network or a wireless network. 
     
     
         18 . The method of  claim 13 , wherein the determined action addresses any one or more optimizations of energy consumption, maintenance, load, resource demand, resource supply, or resource usage. 
     
     
         19 . The method of  claim 13 , wherein the determined action addresses any one or more resource usages selected from: power off, power on, idle state, sleep state, CPU utilization, memory utilization, spectrum allocation or bandwidth allocation. 
     
     
         20 . A method of deploying machine learning models from a network connected cognitive platform, method comprising:
 exchanging instructions, configuration information and data related to at least one machine learning model with an external system;   modifying, translating or interpreting instructions, configuration information exchanged with the external system for use by a cognitive decision maker;   selecting, via a cognitive decision maker, at least one machine learning model from a plurality of machine learning models available and exchange instructions, the configuration information related to deploying the selected machine learning model to the external cognitive device. deploying the selected machine learning model with the external system.   
     
     
         21 . The method of  claim 20 , wherein the cognitive decision maker performing modifications to the selected machine learning model to optimize the performance of the machine learning model based on the hardware platform of the external system. 
     
     
         22 . The method of  claim 20 , further comprising of an asset management module that exchanges instructions and configuration information related to at least one sensor or one actuator to modify the behavior of the external system. 
     
     
         23 . The method of  claim 20 , where the cognitive platform is configured to act as a master or a slave to another cognitive platform. 
     
     
         24 . The method of  claim 20 , wherein the instructions and configuration information exchanged with external cognitive device are related to version information, version selection, version updates, version upgrades, bug fixes, machine learning model modifications, sensors or actuators.