Energy consumption prediction and energy saving control system

Abstract

The present application relates to the technical field of energy consumption control, and particularly relates to an energy consumption prediction and energy saving control system. The technical scheme comprises a data acquisition and communication module, an energy consumption prediction and optimization module, an intelligent control execution module, an energy management and visualization module, and a system coordination mechanism. The data acquisition and communication module is used for collecting operation parameters, environmental parameters and equipment states of building mechanical and electrical equipment in real time, and realizing data aggregation and transmission through a multi-protocol bus. The present application realizes the cooperative optimization of building energy consumption through dynamic prediction and intelligent control, responds to environmental changes in real time, and accurately coordinates the operation of multiple devices, thereby breaking through the limitations of traditional model hysteresis and coupled interference. In combination with multi-source data fusion diagnosis, hidden dangers are actively identified, operation and maintenance costs are reduced, standardized integrated design simplifies construction management, intuitive visual control is provided, and a full-cycle closed loop of energy efficiency improvement and intelligent operation and maintenance is realized.

Description

Technical Field

[0001] This invention relates to the field of energy consumption control technology, and in particular to an energy consumption prediction and energy-saving control system. Background Technology

[0002] Energy efficiency optimization of building automation systems has become a key link in achieving the "dual carbon" goal. However, traditional building energy management systems face the challenge of predictive-control-diagnosis collaborative failure in multi-dimensional dynamic coupling scenarios: the core technical bottleneck is that existing systems cannot simultaneously address the impact of multi-subsystem coupling interference on control accuracy and achieve intelligent multi-source data assessment of equipment energy efficiency status, based on real-time capture of dynamic load disturbances such as personnel density and outdoor environment.

[0003] Specifically, traditional energy consumption prediction models rely on historical data statistics and static load assumptions (such as a fixed thermal coefficient K value and ignoring the dynamic impact of water quality TDS). They lag behind in responding to dynamic scenarios such as instantaneous gatherings of people (leading to a sudden increase in CO2 concentration exceeding 300 ppm) and extreme weather (hourly fluctuations in outdoor temperature and humidity exceeding 15%), resulting in short-term energy consumption prediction error rates generally exceeding 10%. This directly causes the adjustment of operating parameters for air conditioning chiller / heater systems, water pumps, fans, and other equipment to lag behind actual load demands, generating 15%-20% ineffective energy consumption. More importantly, there are strong coupling relationships between subsystems such as air conditioning temperature and humidity control, fresh air volume adjustment, and equipment group control (e.g., changes in fresh air valve opening synchronously affect return air temperature and chilled water load). Traditional control strategies lack multivariate decoupling capabilities (they have not established state-space models for return air temperature, CO2 concentration, and PM2.5 concentration), leading to a "temperature over-adjustment - energy consumption rebound" cycle during control, resulting in a 12%-18% decrease in the system's energy efficiency ratio (EER) compared to the theoretical optimal value. Furthermore, energy efficiency assessment and fault diagnosis rely solely on single-dimensional parameters such as current and voltage (e.g., only monitoring a 10% drop in EER to trigger an early warning), neglecting current harmonic distortion rate (THD>8% increases motor copper loss by 15%) and vibration signals (RMS value of 2-3 m / s in the early stage of bearing wear). 2 The integration analysis of equipment performance with environmental parameters leads to a 3-5 day delay in the discovery of hidden performance losses (such as continuous energy waste caused by insulation aging), and increases operation and maintenance costs by more than 30%.

[0004] The aforementioned problems essentially reflect a technological gap in the traditional system's "dynamic load sensing - coupling interference decoupling - multi-source data decision-making" chain: it cannot dynamically correct the load prediction model using real-time environmental parameters (temperature, humidity, personnel density, water quality indicators), nor can it achieve multi-variable collaborative optimization during control execution, and it lacks the ability to deeply diagnose equipment health status based on harmonic analysis and vibration monitoring. This results in building electromechanical equipment operating in a state of "high energy consumption and low reliability" for extended periods, making it difficult to meet the requirements for reducing energy intensity.

[0005] In summary, this application proposes an energy consumption prediction and energy-saving control system. Summary of the Invention

[0006] The purpose of this invention is to address the problem in the prior art of failing to achieve refined management and efficient energy conservation of building energy consumption, and to propose an energy consumption prediction and energy-saving control system.

[0007] The technical solution of this invention: an energy consumption prediction and energy-saving control system, comprising:

[0008] The data acquisition and communication module is used to collect the operating parameters, environmental parameters and equipment status of building electromechanical equipment in real time, and realize data aggregation and transmission through a multi-protocol bus.

[0009] The energy consumption prediction and optimization module is used to predict building energy consumption trends and generate optimization control commands based on dynamic load calculation, machine learning algorithms and energy efficiency analysis models.

[0010] The intelligent control execution module is used to execute adaptive control strategies, equipment group control logic, and preset energy-saving modes, and dynamically adjust the operating parameters of electromechanical equipment.

[0011] The energy management and visualization module is used for sub-metering of energy consumption, diagnosing equipment faults, and providing a visual monitoring interface.

[0012] The system's collaborative mechanism enables bidirectional data interaction between modules through a real-time database, ensuring that prediction results drive dynamic adjustments to the control strategy.

[0013] Optionally, the data acquisition and communication module includes:

[0014] The distributed sensor unit integrates a temperature sensor, humidity sensor, CO2 concentration sensor, PM2.5 concentration sensor, current transformer, voltage transformer, liquid level sensor, and flow sensor, among which:

[0015] The temperature sensor uses a PT1000 platinum resistance thermometer with an accuracy of ±0.1℃ and is installed in the air conditioning return duct, chilled water supply and return duct, and outdoor environmental monitoring points.

[0016] The current transformer is an open-type Rogowski coil with a range of 0-500A and an accuracy of 0.5 class. It is embedded in the distribution cabinet of the fan, water pump and cold and heat source system.

[0017] The PM2.5 concentration sensor uses the laser scattering principle and has a detection range of 0-1000 μg / m³. 3 Resolution 1μg / m 3 Deployed at the air inlet of the fresh air unit and in key indoor areas;

[0018] The equipment status monitoring unit connects to the fan distribution cabinet, water pump distribution cabinet, lighting distribution box, cold and heat source system, and elevator control cabinet via LONWORKS fieldbus, and collects the equipment's operating status, fault signals, and energy efficiency parameters in real time.

[0019] The fan distribution cabinet has a built-in motor protector that monitors the three-phase current imbalance rate, overload alarm, and insulation resistance value.

[0020] The cold and heat source system controller uploads the chiller unit's evaporator / condenser pressure, compressor operating frequency, and coefficient of performance (COP) via the Modbus protocol;

[0021] The bus communication unit adopts a dual-redundant network architecture, including a LONWORKS fieldbus layer and a TCP / IP Ethernet layer, wherein:

[0022] The LONWORKS bus layer supports free topology, with a transmission rate of 78kbps and a maximum number of nodes of 64, and is used for device-level real-time control signal transmission.

[0023] The TCP / IP Ethernet layer connects data acquisition devices in various areas through a fiber optic ring network. The transmission protocol is BACnet / IP, and it supports OPCUA data subscription and publishing.

[0024] The protocol converter is embedded in the power distribution cabinet controller, converting LONWORKS data packets into JSON format and uploading them to the cloud database via the MQTT protocol.

[0025] Optionally, the energy consumption prediction and optimization module includes:

[0026] The dynamic load calculation unit calculates the cooling load of the air conditioning system in real time based on the chilled water supply and return temperature difference ΔT, flow rate F, and the thermodynamic formula Q=K×F×ΔT, where: K is the chilled water heat coefficient, which is dynamically corrected according to the water quality test report, and the correction formula is K=4186×(1-0.0025×TDS), where TDS is the total dissolved solids content (ppm); historical load data is stored in a time series database, load curves are generated at 15-minute intervals, and periodic features are extracted through a sliding window algorithm;

[0027] The machine learning prediction unit uses the ARIMA model to predict short-term energy consumption trends and combines it with an LSTM neural network to process nonlinear features. Specifically, the input features include outdoor temperature, indoor population density, cumulative equipment operating time, time-of-use electricity price, and historical energy consumption data. The model training uses the Adam optimizer with a learning rate of 0.001, a batch size of 64, and 500 training cycles. The error rate for predicting energy consumption in the next 24 hours is ≤3%.

[0028] The energy efficiency analysis unit assesses the energy efficiency status of a single unit in real time by constructing an equipment energy efficiency ratio (EER) model. The EER calculation formula is EER = cooling capacity (kW) / input power (kW), where the cooling capacity is calculated from the chilled water flow rate and the supply and return water temperature difference. The energy efficiency degradation warning trigger condition is: EER decreases by 10% for 3 consecutive hours or the current harmonic distortion rate exceeds THD>8%.

[0029] Optionally, the intelligent control execution module includes the following units:

[0030] The adaptive control unit uses a fuzzy PID algorithm to adjust the operating parameters of the air conditioning system, specifically including:

[0031] The fuzzy rule base is set to "If the return air temperature deviation is large and the rate of change is fast, then increase the proportional coefficient KP", with a dynamic range of 0.5-2.0 for KP.

[0032] The opening degree of the fresh air valve is controlled in stages according to the CO2 concentration: 30% when CO2 < 800ppm, 50% when CO2-1200ppm, and 100% when CO2 > 1200ppm;

[0033] The frequency of the fan inverter is adjusted by PID according to the deviation between the set value and the actual value of the return air temperature, with an output frequency range of 10-50Hz.

[0034] The equipment group control unit manages the group control of chillers, circulating water pumps, and cooling towers, specifically including:

[0035] The round-robin strategy is based on the cumulative operating time of the equipment. Each time the unit with the shortest operating time is started, the unit with the longest operating time is shut down first.

[0036] The opening of the differential pressure bypass valve is controlled by PID, with a target value of 0.2MPa for the pressure difference between the supply and return water pipes and an adjustment accuracy of ±0.01MPa.

[0037] The small temperature difference compensation technology adjusts the cooling tower fan speed to bring the cooling water return temperature close to the optimal set value of the chiller unit (32℃±0.5℃).

[0038] The energy-saving strategy library stores the following preset control modes:

[0039] Schedule control mode: Divided by weekdays / holidays, the lighting system is set to be fully on from 07:00 to 09:00 and to be switched off between 22:00 and 06:00;

[0040] Infrared sensor control mode: Garage lighting detects traffic flow through a microwave radar sensor, triggers a 30-second delay, and turns off 60 seconds after the vehicle leaves;

[0041] Fire alarm linkage mode: Upon receiving a fire alarm signal, the fresh air valve is forcibly closed, the smoke exhaust valve is opened, and the operation of non-fire pumps is stopped.

[0042] Optionally, the energy management and visualization module includes the following units:

[0043] The energy consumption metering unit uses a 0.2S-level smart meter to sub-meter the energy consumption of the equipment, specifically including:

[0044] Energy consumption of the cold and heat source system is measured by a CT-type electricity meter, with data update interval of 1 minute;

[0045] The energy consumption of the lighting circuit is collected via RS485 bus, which collects the current, voltage and power factor of each circuit.

[0046] The fault diagnosis unit uses an expert system rule engine, with the following rules:

[0047] If the pump current suddenly increases by 20% and lasts for 5 seconds, it is determined to be a stuck rotor fault.

[0048] If the three-phase current imbalance rate of the fan is greater than 15%, a bearing wear warning will be triggered.

[0049] The human-computer interaction unit, based on WebGL technology, constructs a 3D visualization interface, and its specific functions include:

[0050] Dynamically display the water pipe topology diagram of the cold and heat source system, with high temperature alarm points marked in red;

[0051] The energy consumption percentage pie chart is displayed by category, including air conditioning, lighting, water pumps, and elevators, and can be drilled down to the level of each individual equipment item.

[0052] The remote manual intervention interface provides functions for forced start / stop, parameter modification, and policy import. Operation permissions are divided into three levels: administrator, engineer, and visitor.

[0053] Optionally, the system coordination mechanism specifically includes:

[0054] The real-time database uses a time-series database (InfluxDB), with a storage granularity of 1-second raw data. The data partitioning strategy is based on device type, with a data retention period of 30 days for the cold and heat source system and 7 days for the lighting system.

[0055] The output instructions of the energy consumption prediction module are written to the PLC register of the control execution module via the OPCUA protocol. The specific parameters are configured as follows:

[0056] The OPCUA server address is opc.tcp: / / 192.168.1.100:4840, and the session timeout is set to 300 seconds.

[0057] The write register address range is 40001-40050, the data type is Float32, and the encoding rule is Little-Endian.

[0058] After receiving feedback data from the control execution module, the energy management module updates the weights of the energy efficiency analysis model using a sliding window weighted average algorithm, as shown in the formula:

[0059] W new =α·W old +(1-α)·ΔW

[0060] Among them, W new For the updated model weight matrix, W old The model weight matrix before the update is α, which is the forgetting factor (with a value of 0.7-0.9), and ΔW is the model error gradient for the current period. The weight matrix is ​​updated every hour.

[0061] Optionally, the edge computing function of the bus communication unit further includes:

[0062] A lightweight TensorFlow Lite model deployed on a distribution cabinet controller takes equipment current I, voltage V, ambient temperature envTenv, and runtime t as input features, and outputs the load forecast value predPpred for the next 1 hour. The model structure is as follows:

[0063] Input layer: 4 neurons;

[0064] Hidden layers: 2 layers, 8 neurons per layer, with ReLU activation function;

[0065] Output layer: 1 neuron, with a linear activation function;

[0066] The edge node computation cycle is 5 minutes. The prediction results are uploaded to the cloud via the MQTT protocol. The consistency check algorithm uses the root mean square error (RMSE) threshold for judgment, and the formula is as follows:

[0067]

[0068] Among them, P i P represents the load value predicted for the i-th edge node. cloud Let n be the load value predicted by the cloud model in the i-th iteration, n = 12, and n be the total number of data points within the verification period.

[0069] If RMSE > 5%, cloud model resynchronization is triggered; otherwise, edge prediction results are accepted.

[0070] Optionally, the adaptive control unit further includes a multivariable decoupling control unit to resolve the coupling interference between return air temperature, CO2 concentration, and PM2.5 concentration in the air conditioning system, specifically including:

[0071] Establish state-space equations to describe multivariate relationships:

[0072]

[0073] Return air temperature change rate, ΔT: return air temperature deviation, u valve Water valve opening, C CO2 : Indoor CO2 concentration; a1, a2, a3: Temperature dynamic model coefficients; CO2 concentration change rate; u damper : Fresh air valve opening; Q vent : Ventilation volume; b1, b2: CO2 dynamic model coefficients; PM2.5 concentration change rate; u filter c1, c2: Filter efficiency; c1, c2: PM2.5 dynamic model coefficients;

[0074] A feedforward-feedback composite control system is adopted. The feedforward controller pre-adjusts the opening of the fresh air valve based on changes in CO2 concentration, while the feedback controller adjusts the opening of the water valve based on the return air temperature deviation. The control rate is:

[0075]

[0076] Among them, e CO2 =C CO2,set -C CO2 CO2 concentration deviation, e T =T set -T return : Return air temperature

[0077] Degree deviation; K p K i K d : K is the PID parameter for CO2 concentration control. p ′, K i ′, K d ′: PID parameters for temperature control, dynamically adjusted using fuzzy rules.

[0078] Optionally, the fault diagnosis unit further includes a harmonic distortion rate analysis unit for detecting motor faults, specifically including:

[0079] Acquire the current signal and perform a Fast Fourier Transform (FFT) to calculate the Total Harmonic Distortion (THD). The formula is as follows:

[0080]

[0081] Where I1 is the effective value of the fundamental current, I h The effective value of the h-th harmonic current is given by , and THD is the total harmonic distortion rate.

[0082] Optional, set fault determination rules:

[0083] If THD > 8% and continues for 10 minutes, the "motor winding insulation aging" alarm will be triggered.

[0084] If the third harmonic component I3 / I1 > 5% and the fifth harmonic component I5 / I1 > 3%, then it is determined to be "power supply voltage imbalance".

[0085] Based on vibration sensor data, when THD exceeds the limit and the vibration acceleration RMS value is >4 m / s², 2 At that time, it was determined to be a "combined fault of bearing wear and electrical fault".

[0086] Compared with the prior art, this application includes at least one of the following beneficial technical effects:

[0087] By collecting multi-dimensional equipment and environmental data in real time, and combining dynamic load forecasting and machine learning algorithms, the real-time performance and accuracy of energy consumption forecasting are significantly improved, overcoming the limitations of traditional models that rely on static assumptions.

[0088] By employing multivariable decoupling control and adaptive strategies, the coupling relationships of subsystems such as air conditioning, lighting, and ventilation are effectively coordinated, thereby achieving global energy efficiency optimization for equipment group control.

[0089] By integrating harmonic analysis, vibration monitoring, and environmental parameter fusion diagnosis, it proactively identifies hidden equipment faults and energy efficiency degradation issues, reducing the passivity of operation and maintenance;

[0090] Standardized integrated design of strong and weak current systems simplifies the construction process and reduces the need for professional coordination, while the visualization platform provides intuitive energy management and remote control capabilities.

[0091] This invention optimizes building energy consumption through dynamic prediction and intelligent control, responds to environmental changes in real time, and precisely coordinates the operation of multiple devices, overcoming the limitations of traditional models in terms of lag and coupling interference. It also proactively identifies potential hazards by combining multi-source data fusion diagnosis, reducing operation and maintenance costs. At the same time, standardized integrated design simplifies construction management, provides intuitive and visual control, and realizes a closed loop of energy efficiency improvement and intelligent operation and maintenance throughout the entire life cycle. Attached Figure Description

[0092] Figure 1 This is a block diagram illustrating the principle of an energy consumption prediction and energy-saving control system. Detailed Implementation

[0093] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0094] Example

[0095] I. For example Figure 1As shown, the energy consumption prediction and energy-saving control system proposed in this invention includes a data acquisition and communication module, an energy consumption prediction and optimization module, an intelligent control execution module, an energy management and visualization module, and a system coordination mechanism. Each part will be described in detail below.

[0096] II. The data acquisition and communication module is used to collect real-time operating parameters, environmental parameters, and equipment status of building electromechanical equipment, and to achieve data aggregation and transmission through a multi-protocol bus; specifically including:

[0097] The distributed sensor unit integrates a temperature sensor, humidity sensor, CO2 concentration sensor, PM2.5 concentration sensor, current transformer, voltage transformer, liquid level sensor, and flow sensor, among which:

[0098] The temperature sensor uses a PT1000 platinum resistance thermometer with an accuracy of ±0.1℃ and is installed in the air conditioning return duct, chilled water supply and return duct, and outdoor environmental monitoring points.

[0099] The current transformer is an open-type Rogowski coil with a range of 0-500A and an accuracy of 0.5 class. It is embedded in the distribution cabinet of the fan, water pump and cold and heat source system.

[0100] The PM2.5 concentration sensor uses the laser scattering principle and has a detection range of 0-1000 μg / m³. 3 Resolution 1μg / m 3 Deployed at the air inlet of the fresh air unit and in key indoor areas;

[0101] The equipment status monitoring unit connects to the fan distribution cabinet, water pump distribution cabinet, lighting distribution box, cold and heat source system, and elevator control cabinet via LONWORKS fieldbus, and collects the equipment's operating status, fault signals, and energy efficiency parameters in real time.

[0102] The fan distribution cabinet has a built-in motor protector that monitors the three-phase current imbalance rate, overload alarm, and insulation resistance value.

[0103] The cold and heat source system controller uploads the chiller unit's evaporator / condenser pressure, compressor operating frequency, and coefficient of performance (COP) via the Modbus protocol;

[0104] The bus communication unit adopts a dual-redundant network architecture, including a LONWORKS fieldbus layer and a TCP / IP Ethernet layer, wherein:

[0105] The LONWORKS bus layer supports free topology, with a transmission rate of 78kbps and a maximum number of nodes of 64, and is used for device-level real-time control signal transmission.

[0106] The TCP / IP Ethernet layer connects data acquisition devices in various areas through a fiber optic ring network. The transmission protocol is BACnet / IP, and it supports OPCUA data subscription and publishing.

[0107] The protocol converter is embedded in the power distribution cabinet controller, converting LONWORKS data packets into JSON format and uploading them to the cloud database via the MQTT protocol.

[0108] It should be noted that the edge computing functions of the bus communication unit include:

[0109] A lightweight TensorFlow Lite model deployed on a distribution cabinet controller takes equipment current I, voltage V, ambient temperature envTenv, and runtime t as input features, and outputs the load forecast value predPpred for the next 1 hour. The model structure is as follows:

[0110] Input layer: 4 neurons;

[0111] Hidden layers: 2 layers, 8 neurons per layer, with ReLU activation function;

[0112] Output layer: 1 neuron, with a linear activation function;

[0113] The edge node computation cycle is 5 minutes. The prediction results are uploaded to the cloud via the MQTT protocol. The consistency check algorithm uses the root mean square error (RMSE) threshold for judgment, and the formula is as follows:

[0114]

[0115] Among them, P i P represents the load value predicted for the i-th edge node. cloud Let n be the load value predicted by the cloud model in the i-th iteration, n = 12, and n be the total number of data points within the verification period.

[0116] If RMSE > 5%, cloud model resynchronization is triggered; otherwise, edge prediction results are accepted.

[0117] The data acquisition and communication module integrates multiple types of sensors (temperature, current, environmental quality, etc.) and distributed monitoring units to achieve comprehensive real-time acquisition of operating parameters, environmental conditions, and equipment fault signals of building electromechanical equipment, solving the problems of data silos and protocol compatibility in traditional systems. A dual-redundant network architecture ensures the stability and real-time performance of data transmission, providing multi-dimensional, high-precision basic data support for subsequent modules and ensuring accurate perception of equipment status and environmental changes.

[0118] III. In this embodiment, the energy consumption prediction and optimization module is used to predict building energy consumption trends and generate optimization control commands based on dynamic load calculation, machine learning algorithms, and energy efficiency analysis models; specifically including:

[0119] The dynamic load calculation unit calculates the cooling load of the air conditioning system in real time based on the chilled water supply and return temperature difference ΔT, flow rate F, and the thermodynamic formula Q=K×F×ΔT, where: K is the chilled water heat coefficient, which is dynamically corrected according to the water quality test report, and the correction formula is K=4186×(1-0.0025×TDS), where TDS is the total dissolved solids content (ppm); historical load data is stored in a time series database, load curves are generated at 15-minute intervals, and periodic features are extracted through a sliding window algorithm;

[0120] The machine learning prediction unit uses the ARIMA model to predict short-term energy consumption trends and combines it with an LSTM neural network to process nonlinear features. Specifically, the input features include outdoor temperature, indoor population density, cumulative equipment operating time, time-of-use electricity price, and historical energy consumption data. The model training uses the Adam optimizer with a learning rate of 0.001, a batch size of 64, and 500 training cycles. The error rate for predicting energy consumption in the next 24 hours is ≤3%.

[0121] The energy efficiency analysis unit assesses the energy efficiency status of a single unit in real time by constructing an equipment energy efficiency ratio (EER) model. The EER calculation formula is EER = cooling capacity (kW) / input power (kW), where the cooling capacity is calculated from the chilled water flow rate and the supply and return water temperature difference. The energy efficiency degradation warning trigger condition is: EER decreases by 10% for 3 consecutive hours or the current harmonic distortion rate exceeds THD>8%.

[0122] In this embodiment, based on dynamic load calculation and machine learning algorithms, the limitations of traditional static models are overcome. This allows for real-time capture of the impact of dynamic factors such as population density and outdoor environment on energy consumption, accurately predicting building energy consumption trends. Combined with an energy efficiency analysis model, the energy efficiency status of equipment is evaluated in real time, generating optimized control commands. This provides a scientific basis for intelligent regulation, achieving a leap from "historical data statistics" to "dynamic trend prediction," and enhancing the foresight and targeting of energy management.

[0123] IV. Intelligent Control Execution Module, used to execute adaptive control strategies, equipment group control logic, and preset energy-saving modes, dynamically adjusting the operating parameters of electromechanical equipment; the intelligent control execution module includes the following units:

[0124] The adaptive control unit uses a fuzzy PID algorithm to adjust the operating parameters of the air conditioning system, specifically including:

[0125] The fuzzy rule base is set to "If the return air temperature deviation is large and the rate of change is fast, then increase the proportional coefficient KP", with a dynamic range of 0.5-2.0 for KP.

[0126] The opening degree of the fresh air valve is controlled in stages according to the CO2 concentration: 30% when CO2 < 800ppm, 50% when CO2-1200ppm, and 100% when CO2 > 1200ppm;

[0127] The frequency of the fan inverter is adjusted by PID according to the deviation between the set value and the actual value of the return air temperature, with an output frequency range of 10-50Hz.

[0128] The adaptive control unit further includes a multivariable decoupling control unit to resolve the coupling interference between return air temperature, CO2 concentration, and PM2.5 concentration in the air conditioning system, specifically including:

[0129] Establish state-space equations to describe multivariate relationships:

[0130]

[0131] Return air temperature change rate, ΔT: return air temperature deviation, u valve Water valve opening, C CO2 : Indoor CO2 concentration; a1, a2, a3: Temperature dynamic model coefficients; CO2 concentration change rate; u damper : Fresh air valve opening; Q vent : Ventilation volume; b1, b2: CO2 dynamic model coefficients; PM2.5 concentration change rate; u filter c1, c2: Filter efficiency; c1, c2: PM2.5 dynamic model coefficients;

[0132] A feedforward-feedback composite control system is adopted. The feedforward controller pre-adjusts the opening of the fresh air valve based on changes in CO2 concentration, while the feedback controller adjusts the opening of the water valve based on the return air temperature deviation. The control rate is:

[0133]

[0134] Among them, e CO2 =C CO2,set -C CO2 CO2 concentration deviation, e T =T set -T return Return air temperature deviation; K p K i K d : K is the PID parameter for CO2 concentration control. p ′, K i ′, K d ′: PID parameters for temperature control, dynamically adjusted using fuzzy rules.

[0135] The equipment group control unit manages the group control of chillers, circulating water pumps, and cooling towers, specifically including:

[0136] The round-robin strategy is based on the cumulative operating time of the equipment. Each time the unit with the shortest operating time is started, the unit with the longest operating time is shut down first.

[0137] The opening of the differential pressure bypass valve is controlled by PID, with a target value of 0.2MPa for the pressure difference between the supply and return water pipes and an adjustment accuracy of ±0.01MPa.

[0138] The small temperature difference compensation technology adjusts the cooling tower fan speed to bring the cooling water return temperature close to the optimal set value of the chiller unit (32℃±0.5℃).

[0139] The energy-saving strategy library stores the following preset control modes:

[0140] Schedule control mode: Divided by weekdays / holidays, the lighting system is set to be fully on from 07:00 to 09:00 and to be switched off between 22:00 and 06:00;

[0141] Infrared sensor control mode: Garage lighting detects traffic flow through a microwave radar sensor, triggers a 30-second delay, and turns off 60 seconds after the vehicle leaves;

[0142] Fire alarm linkage mode: Upon receiving a fire alarm signal, the fresh air valve is forcibly closed, the smoke exhaust valve is opened, and the operation of non-fire pumps is stopped.

[0143] In this embodiment, adaptive control algorithms (such as fuzzy PID and multivariable decoupling) and equipment group control strategies effectively solve the problem of coupling interference between multiple subsystems, enabling dynamic adjustment and coordinated optimization of operating parameters for equipment such as air conditioning, ventilation, and lighting. Preset energy-saving modes (timetable control, sensor control, fire alarm linkage, etc.) automatically adapt to different scenario requirements, reducing ineffective energy consumption while ensuring environmental comfort, and promoting the transformation of equipment from "independent control" to "global intelligent collaboration."

[0144] V. In this embodiment, the energy management and visualization module is used for itemized energy consumption metering, diagnosing equipment faults, and providing a visual monitoring interface. It utilizes high-precision smart meters to achieve itemized energy consumption metering, and combines an expert system rule engine with multi-source data fusion analysis to accurately diagnose equipment faults (such as bearing wear and winding aging) and provide early warnings of energy efficiency degradation. The three-dimensional visualization interface presents real-time energy consumption distribution, equipment operating status, and fault location, supporting drill-down analysis and remote intervention from the system level to the equipment level, improving the precision and efficiency of operation and maintenance management, and achieving integrated management of "data visualization—fault diagnosis—strategy intervention." The energy management and visualization module includes the following units:

[0145] The energy consumption metering unit uses a 0.2S-level smart meter to sub-meter the energy consumption of the equipment, specifically including: the energy consumption of the cold and heat source system is measured by a CT-type power meter, with a data update interval of 1 minute;

[0146] The energy consumption of the lighting circuit is collected via RS485 bus, which collects the current, voltage and power factor of each circuit.

[0147] The fault diagnosis unit employs an expert system rule engine with the following rules: If the pump current suddenly increases by 20% and lasts for 5 seconds, it is determined to be a blade jamming fault; if the three-phase current imbalance rate of the fan is >15%, a bearing wear warning is triggered; the fault diagnosis unit further includes a harmonic distortion rate analysis unit for detecting motor faults, specifically including:

[0148] Acquire the current signal and perform a Fast Fourier Transform (FFT) to calculate the Total Harmonic Distortion (THD). The formula is as follows:

[0149]

[0150] Where I1 is the effective value of the fundamental current, I h The effective value of the h-th harmonic current is given by , and THD is the total harmonic distortion rate.

[0151] Among them, the fault determination rules are set as follows:

[0152] If THD > 8% and continues for 10 minutes, the "motor winding insulation aging" alarm will be triggered.

[0153] If the third harmonic component I3 / I1 > 5% and the fifth harmonic component I5 / I1 > 3%, then it is determined to be "power supply voltage imbalance".

[0154] Based on vibration sensor data, when THD exceeds the limit and the vibration acceleration RMS value is >4 m / s², 2 At that time, it was determined to be a "combined fault of bearing wear and electrical fault".

[0155] The human-computer interaction unit, based on WebGL technology, constructs a 3D visualization interface, and its specific functions include:

[0156] Dynamically display the water pipe topology diagram of the cold and heat source system, with high temperature alarm points marked in red;

[0157] The energy consumption percentage pie chart is displayed by category, including air conditioning, lighting, water pumps, and elevators, and can be drilled down to the level of each individual equipment item.

[0158] The remote manual intervention interface provides functions for forced start / stop, parameter modification, and policy import. Operation permissions are divided into three levels: administrator, engineer, and visitor.

[0159] VI. The system's collaborative mechanism achieves bidirectional data interaction between modules through a real-time database, ensuring that prediction results drive dynamic adjustments to the control strategy. Specifically, this includes:

[0160] The real-time database uses a time-series database (InfluxDB), with a storage granularity of 1-second raw data. The data partitioning strategy is based on device type, with a data retention period of 30 days for the cold and heat source system and 7 days for the lighting system.

[0161] The output instructions of the energy consumption prediction module are written to the PLC register of the control execution module via the OPCUA protocol. The specific parameters are configured as follows:

[0162] The OPCUA server address is opc.tcp: / / 192.168.1.100:4840, and the session timeout is set to 300 seconds.

[0163] The write register address range is 40001-40050, the data type is Float32, and the encoding rule is Little-Endian.

[0164] After receiving feedback data from the control execution module, the energy management module updates the weights of the energy efficiency analysis model using a sliding window weighted average algorithm, as shown in the formula:

[0165] W new =α·W old +(1-α)·ΔW

[0166] Among them, W new For the updated model weight matrix, W old The model weight matrix before the update is α, which is the forgetting factor (with a value of 0.7-0.9), and ΔW is the model error gradient for the current period. The weight matrix is ​​updated every hour.

[0167] In this embodiment, a fully closed-loop collaborative system of "data acquisition—predictive analysis—control execution—feedback optimization" is constructed through a real-time database and a two-way data interaction protocol to ensure real-time data sharing and dynamic linkage among modules. The collaborative verification mechanism of edge computing and cloud models improves local data processing efficiency while ensuring the consistency of prediction results, enhances the system's rapid response capability and overall robustness to complex scenarios, and forms a highly efficient operating mode of "prediction-driven control and control feedback optimization".

[0168] This invention provides a complete chain from data perception and predictive analysis to intelligent control and visual management, breaking through the bottlenecks of traditional systems in data integration, prediction accuracy, control coordination, and operation and maintenance efficiency. It enables refined management of building energy consumption, systematic improvement of equipment energy efficiency, and intelligent transformation of operation and maintenance decision-making, providing solid technical support for green building and high-efficiency energy conservation.

[0169] The above specific embodiments are merely several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. An energy consumption prediction and energy-saving control system, characterized in that, include: The data acquisition and communication module is used to collect real-time operating parameters, environmental parameters, and equipment status of building electromechanical equipment, and to achieve data aggregation and transmission through a multi-protocol bus; the data acquisition and communication module specifically includes: The distributed sensor unit integrates a temperature sensor, humidity sensor, CO2 concentration sensor, PM2.5 concentration sensor, current transformer, voltage transformer, liquid level sensor, and flow sensor. The equipment status monitoring unit connects to the fan distribution cabinet, water pump distribution cabinet, lighting distribution box, cold and heat source system, and elevator control cabinet via LONWORKS fieldbus to collect real-time equipment operating status, fault signals, and energy efficiency parameters. The bus communication unit adopts a dual-redundant network architecture, including a LONWORKS fieldbus layer and a TCP / IP Ethernet layer; The edge computing function of the bus communication unit includes: A lightweight TensorFlow Lite model deployed on a distribution cabinet controller takes as input features equipment current I, voltage V, ambient temperature envTenv, and runtime t, and outputs the load forecast value predP pred for the next hour. The model structure is as follows: Input layer: 4 neurons; Hidden layers: 2 layers, 8 neurons per layer, with ReLU activation function; Output layer: 1 neuron, with a linear activation function; The edge node computation cycle is 5 minutes. The prediction results are uploaded to the cloud via the MQTT protocol. The consistency check algorithm uses the root mean square error threshold for judgment, and the formula is: ； in, For the first Predicted load values ​​for secondary edge nodes; For the first Load values ​​predicted by the cloud model , which is the total number of data points within the verification period. If RMSE > 5%, cloud model resynchronization is triggered; otherwise, edge prediction results are accepted. The energy consumption prediction and optimization module is used to predict building energy consumption trends and generate optimized control commands based on dynamic load calculation, machine learning algorithms, and energy efficiency analysis models. The energy consumption prediction and optimization module includes: The dynamic load calculation unit calculates the cooling load of the air conditioning system in real time based on the chilled water supply and return temperature difference ΔT, flow rate F, and the thermodynamic formula Q=K×F×ΔT, where K is the chilled water heat coefficient, which is dynamically corrected according to the water quality test report, and the correction formula is K=4186×(1-0.0025×TDS), where TDS is the total dissolved solids content; historical load data is stored in a time series database, load curves are generated at 15-minute intervals, and periodic features are extracted through a sliding window algorithm; The machine learning prediction unit uses the ARIMA model to predict short-term energy consumption trends and combines it with an LSTM neural network to process nonlinear features. The energy efficiency analysis unit assesses the energy efficiency status of individual devices in real time by constructing an energy efficiency ratio model for the equipment. The intelligent control execution module is used to execute adaptive control strategies, equipment group control logic, and preset energy-saving modes, and dynamically adjust the operating parameters of electromechanical equipment. The energy management and visualization module is used for sub-metering of energy consumption, diagnosing equipment faults, and providing a visual monitoring interface. The system coordination mechanism, through a real-time database, enables bidirectional data interaction between modules, ensuring that prediction results drive dynamic adjustments to the control strategy; the system coordination mechanism specifically includes: The real-time database adopts a time-series database, with the storage granularity set to 1-second level raw data. The data partitioning strategy is divided according to equipment type, with the data retention period for cold and heat source systems being 30 days and the data retention period for lighting systems being 7 days. The output instructions of the energy consumption prediction and optimization module are written into the PLC register of the control execution module via the OPC UA protocol; After receiving feedback data from the control execution module, the energy management and visualization module updates the weights of the energy efficiency analysis model using a sliding window weighted average algorithm, as shown in the formula: ； in, This is the updated model weight matrix. This is the model weight matrix before the update. Forgetting factor, The weight matrix is ​​updated hourly to represent the model error gradient for the current period.

2. The energy consumption prediction and energy-saving control system according to claim 1, characterized in that, The intelligent control execution module includes the following units: The adaptive control unit uses a fuzzy PID algorithm to adjust the operating parameters of the air conditioning system. The equipment group control unit controls the chiller, circulating water pump and cooling tower in groups; The energy-saving strategy library stores the following preset control modes: Schedule control mode: Divided by weekdays / holidays, the lighting system is set to be fully on from 07:00 to 09:00 and to be switched off between 22:00 and 06:00; Infrared sensor control mode: Garage lighting detects traffic flow through a microwave radar sensor, triggers a 30-second delay, and turns off 60 seconds after the vehicle leaves; Fire alarm linkage mode: Upon receiving a fire alarm signal, the fresh air valve is forcibly closed, the smoke exhaust valve is opened, and the operation of non-fire pumps is stopped.

3. The energy consumption prediction and energy-saving control system according to claim 1, characterized in that, The energy management and visualization module includes the following units: The energy consumption metering unit uses a 0.2S-level smart meter to sub-meter the energy consumption of the equipment; The fault diagnosis unit uses an expert system rule engine, with the following rules: If the pump current suddenly increases by 20% and lasts for 5 seconds, it is determined to be a stuck rotor fault. If the three-phase current imbalance rate of the fan is greater than 15%, a bearing wear warning will be triggered. The human-computer interaction unit uses WebGL technology to build a 3D visualization interface.

4. The energy consumption prediction and energy-saving control system according to claim 2, characterized in that, The adaptive control unit further includes a multivariable decoupling control unit, used to resolve the coupling interference between return air temperature, CO2 concentration, and PM2.5 concentration in the air conditioning system, specifically including: Establish state-space equations to describe multivariate relationships: ； in, Return air temperature change rate Return air temperature deviation Water valve opening, Indoor CO2 concentration; , , Temperature dynamic model coefficients; : Rate of change in CO2 concentration; Fresh air valve opening degree; Ventilation volume; , : CO2 dynamic model coefficients; PM2.5 concentration change rate; Filter efficiency; , PM2.5 dynamic model coefficients; A feedforward-feedback composite control system is adopted. The feedforward controller pre-adjusts the opening of the fresh air valve based on changes in CO2 concentration, while the feedback controller adjusts the opening of the water valve based on the return air temperature deviation. The control law is as follows: in, CO2 concentration ; ； degree deviation, Return air temperature deviation; , , : These are the PID parameters for CO2 concentration control. , , The PID parameters for temperature control are dynamically adjusted using fuzzy rules.

5. The energy consumption prediction and energy-saving control system according to claim 3, characterized in that, The fault diagnosis unit further includes a harmonic distortion rate analysis unit for detecting motor faults, specifically including: The current signal is acquired and subjected to a Fast Fourier Transform (FFT). The total harmonic distortion (THD) is calculated using the following formula: ； in, This is the effective value of the fundamental current. For the first RMS value of subharmonic current This represents the total harmonic distortion rate.

6. The energy consumption prediction and energy-saving control system according to claim 5, characterized in that, Set fault determination rules: If THD > 8% and continues for 10 minutes, an alarm for "motor winding insulation aging" will be triggered. If the third harmonic component / >5% and the 5th harmonic component / If the value is greater than 3%, it is determined to be "power supply voltage imbalance"; Based on vibration sensor data, when the THD exceeds the limit and the vibration acceleration RMS value is >4m / s², it is determined to be a "combined fault of bearing wear and electrical fault".

Images