Home energy management system and energy saving decision method thereof

Abstract

An energy-saving decision-making method for a home energy management system includes: acquiring multiple sets of environmental data and multiple sets of electricity consumption data, each set of environmental data including ambient temperature and humidity, and each set of electricity consumption data including the power consumption of each appliance; converting the multiple sets of environmental data into multiple sets of semantic variables to establish fuzzy logic rules based on the multiple sets of semantic variables; performing an optimization algorithm based on a reward function to find a set of optimal environmental data for optimizing the reward function from the multiple sets of environmental data; and performing fuzzy logic inference based on the fuzzy logic rules to generate a set of optimal decision parameters for controlling multiple appliances using the set of optimal environmental data.

Claims

1. An energy-saving decision-making method for a home energy management system, executed by a computer device, comprising: Obtain complex arrays of environmental data and complex arrays of power consumption data, wherein each set of the complex arrays of environmental data includes an ambient temperature and an ambient humidity, and each set of the complex arrays of power consumption data includes the power consumption of each of a plurality of electrical appliances; convert the complex arrays of environmental data into complex arrays of semantic variables, and establish a fuzzy logic rule based on the complex arrays of semantic variables; execute an optimization algorithm based on a reward function to find a set of optimal environmental data that optimizes the reward function from the complex arrays of environmental data; and perform fuzzy logic inference based on the fuzzy logic rule to generate a set of optimal decision parameters for controlling these electrical appliances using the set of optimal environmental data; wherein these electrical appliances are all IoT-enabled devices and consist of a dehumidifier, an air conditioner, and a fan, wherein the set of optimal decision parameters includes a set temperature, an airflow strength, and an operating mode of the air conditioner, an on / off state and a set humidity of the dehumidifier, and an on / off state of the fan; wherein the reward function is expressed as follows: , Here, is a temperature difference and is associated with the ambient temperature and a user-preferred temperature dataset; is a humidity difference and is associated with the ambient humidity and a user-preferred humidity dataset; and are the power consumption of the air conditioner, the dehumidifier, and the fan, respectively; and are the operating durations of the air conditioner, the dehumidifier, and the fan, respectively; is a temperature weight associated with ; is a humidity weight associated with ; is the electricity price during the electricity usage period.

2. The energy-saving decision-making method for a home energy management system as described in claim 1, further comprising: The reward function is used to calculate a reward value by substituting a set of environmental data from the complex array of environmental data and a set of electricity consumption data from the corresponding complex array of electricity consumption data; and the optimization algorithm is executed with the goal of optimizing the reward value to the minimum value.

3. The energy-saving decision-making method for a home energy management system as described in claim 1 further includes: The absolute value of thermal comfort is not greater than a threshold as a constraint condition for the optimization algorithm, so that the optimization algorithm also applies a penalty based on the constraint condition.

4. The energy-saving decision-making method for a home energy management system as described in claim 3, wherein thermal comfort is indicated by predicted mean vote (PMV).

5. The energy-saving decision-making method for a home energy management system as described in claim 1, wherein when the ambient temperature is less than a lower limit of the user-preferred temperature dataset, = the lower limit - the ambient temperature, and; when the ambient temperature is greater than a higher limit of the user-preferred temperature dataset, = the ambient temperature - the higher limit, and; and when the ambient temperature is neither less than the lower limit nor greater than the higher limit, ...

6. The energy-saving decision-making method for a home energy management system as described in claim 1, wherein when the ambient humidity is less than a lower humidity limit of the user-preferred humidity dataset, = the lower humidity limit - the ambient humidity, and; when the ambient humidity is greater than a higher humidity limit of the user-preferred humidity dataset, = the ambient humidity - the higher humidity limit, and; and when the ambient humidity is neither less than the lower humidity limit nor greater than the higher humidity limit, ...

7. The energy-saving decision-making method for a home energy management system as described in claim 1, wherein the fuzzy logic inference employs an adaptive neuro fuzzy inference system (ANFIS).

8. The energy-saving decision-making method for a home energy management system as described in claim 1, further comprising: Calculate one unit of electricity based on the optimal decision parameters of this set; And based on the amount of electricity saved, determine whether to re-execute the optimization algorithm.

9. A home energy management system, comprising: A sensor for sensing complex arrays of environmental data, each set of which includes an ambient temperature and an ambient humidity; a plurality of electrical appliances; and a computer device communicatively connected to the sensor to acquire the complex arrays of environmental data from the sensor, and communicatively connected to the electrical appliances to acquire complex arrays of power consumption data from the electrical appliances, each set of the complex arrays of power consumption data including a power consumption of each of the electrical appliances, wherein the computer device is configured to perform the following steps: converting the complex arrays of environmental data into complex arrays of semantic variables to establish a fuzzy logic rule based on the complex arrays of semantic variables; performing an optimization algorithm based on a reward function to find a set of optimal environmental data that optimizes the reward function from the complex arrays of environmental data; and performing a fuzzy logic inference based on the fuzzy logic rule to generate a set of optimal decision parameters for controlling the electrical appliances using the set of optimal environmental data; These appliances are all IoT-enabled devices, consisting of a dehumidifier, an air conditioner, and a fan. The optimal decision parameters include a set temperature, airflow strength, and operating mode for the air conditioner; an on / off state and a set humidity for the dehumidifier; and an on / off state for the fan. The reward function is expressed as follows: where is a temperature difference associated with the ambient temperature and a user-preferred set temperature dataset; is a humidity difference associated with the ambient humidity and a user-preferred set humidity dataset; and are the power consumption of the air conditioner, dehumidifier, and fan, respectively; and are the operating durations of the air conditioner, dehumidifier, and fan, respectively; is a temperature weight associated with ; is a humidity weight associated with ; and is the electricity price during the usage period.

10. The home energy management system as claimed in claim 9, wherein the computer device is further configured to perform the following steps: substituting a set of environmental data from the complex array of environmental data and a set of electricity consumption data from the corresponding complex array of electricity consumption data into the reward function to calculate a reward value; and performing the optimization algorithm with the objective of optimizing the reward value to have a minimum value.

11. The home energy management system as claimed in claim 9, wherein the computer device is further configured to perform the following steps: using an absolute value of thermal comfort not greater than a threshold as a constraint on the optimization algorithm, such that the optimization algorithm also applies a penalty based on the constraint.

12. The home energy management system as described in claim 11, wherein the thermal comfort is measured using a predicted average vote.

13. The home energy management system as claimed in claim 9, wherein when the ambient temperature is less than a lower limit of the user-preferred temperature dataset, = the lower limit - the ambient temperature, and; when the ambient temperature is greater than a higher limit of the user-preferred temperature dataset, = the ambient temperature - the higher limit, and; and when the ambient temperature is neither less than the lower limit nor greater than the higher limit, ...

14. The home energy management system as claimed in claim 9, wherein when the ambient humidity is less than a lower humidity limit of the user-preferred humidity dataset, = the lower humidity limit - the ambient humidity, and; when the ambient humidity is greater than a higher humidity limit of the user-preferred humidity dataset, = the ambient humidity - the higher humidity limit, and; and when the ambient humidity is neither less than the lower humidity limit nor greater than the higher humidity limit, ...

15. The home energy management system as described in claim 9, wherein the fuzzy logic inference employs an adaptive neurofuzzy inference system.

16. The home energy management system as claimed in claim 9, wherein the computer device is further configured to perform the following steps: calculate a power saving based on the set of optimal decision parameters; and determine whether to re-execute the optimization algorithm based on the power saving.

Images