A Distributed Method and System for Monitoring the Condition of Poultry House Fans
By adopting a three-layer distributed architecture and a manual-adaptive dual closed-loop threshold management system, the reliability and scalability issues of poultry house fan status monitoring have been resolved, enabling accurate monitoring of fan status and fault early warning, and reducing false alarm rate and equipment maintenance costs.
Patent Information
- Application Number
- CN202611484345.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-22
- Publication Date
- 2026-11-17
AI Technical Summary
Existing poultry house environmental control systems suffer from several problems, including unreliable reliance on manual inspections for fan status monitoring, weak scalability of centralized detection, inability of a single threshold to adapt to multiple fan specifications, and high false alarm rate.
Adopting a three-layer distributed architecture, fault determination is decentralized to the remote current detection module. Combined with the wind turbine control commands broadcast by the main control and multi-dimensional current characteristics, the system achieves accurate monitoring of wind turbine operating status and fault early warning through manual-adaptive dual closed-loop threshold management.
It achieves accurate monitoring of wind turbine status, reduces false alarm rate, adapts to multiple wind turbine specifications, reduces equipment maintenance costs, and improves system scalability and anti-interference capability.
Smart Images

Figure CN123110124A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent control technology for poultry house environment, and in particular to a method and system for monitoring the status of poultry house fans based on a distributed system. This system can detect the operating status of each ventilation fan in real time and issue timely alarms to ensure ventilation safety in poultry houses. Background Technology
[0002] Fans are the core equipment for ventilation and temperature control in poultry houses, and their stable operation directly determines the safety of the breeding environment. Existing poultry house environmental control systems have many shortcomings in monitoring fan status: First, the environmental control terminal is located at the front of the poultry house, while the fans are distributed at the rear. After the main controller issues the start / stop control command for the fans, it is impossible to obtain feedback on the actual operation of the fans. The status of the fans can only be judged by the farmers through visual inspection and listening. The manual inspection cycle is long, and the inspection results are greatly affected by the experience of the personnel, the size of the farm, and day and night weather conditions, resulting in a high risk of missed inspections and misjudgments.
[0003] Secondly, poultry house configurations vary greatly, and the number of fan configurations is not fixed. Traditional centralized current detection equipment has a fixed number of hardware detection channels and poor scalability; once the centralized detection unit fails, all fan monitoring will fail, requiring the entire unit to be replaced, resulting in high equipment maintenance costs.
[0004] Third, multiple types of fans with varying power ratings may be deployed within the same factory area or even the same poultry house, with significant differences in rated current between the different fans. Most existing monitoring solutions use a fixed single current threshold, which cannot adapt to scenarios where multiple fan specifications are used together; after long-term aging of the fans, the operating current will drift, and fixed thresholds are prone to false alarms or missed alarms.
[0005] Some existing distributed motor data acquisition solutions only collect raw current data at the acquisition nodes and upload it all to the main control unit, with all fault diagnosis calculations handled centrally by the main control unit. This solution involves a large amount of data transmitted via the bus, placing a heavy burden on the main control unit's computing power. The remote acquisition nodes cannot obtain the raw control commands for the fans issued by the environmental control unit and rely solely on current values for judgment. The transient impacts during fan start-up and shutdown can easily lead to false alarms, making it unsuitable for use in poultry houses with multiple fans and harsh operating conditions.
[0006] In summary, there is an urgent need for a distributed status monitoring solution for fans that is adapted to the unique working conditions of poultry houses, can be flexibly expanded, supports multiple fan specifications, and reduces false alarm rates. Summary of the Invention
[0007] The purpose of this invention is to overcome the shortcomings of existing technologies, such as poor reliability of manual inspection, weak scalability of centralized detection, inability of single threshold values to adapt to multiple fan specifications, and high false alarm rate. This invention provides a distributed method and system for monitoring the status of poultry house fans. Addressing the challenges of mixed installations of multiple fan specifications, flexible addition and removal of fan nodes, and high incidence of belt-driven mechanical failures in poultry houses, this invention employs a three-layer distributed architecture. Fault determination is decentralized to a remote current detection module, combined with fan control commands broadcast by the main control unit, multi-dimensional current characteristics, and a manual-adaptive dual closed-loop threshold management system, to achieve accurate monitoring of fan operating status and fault early warning.
[0008] To achieve the above objectives, the technical solution of the present invention is implemented as follows: A distributed poultry house fan status monitoring method is implemented through a three-layer distributed framework. The three-layer distributed framework includes: multiple current detection modules, a current integrated communication module, and a main control module. Each current detection module corresponds one-to-one with a poultry house fan.
[0009] The first layer: the current detection module, responsible for sampling the current of a single wind turbine, digital filtering, multi-dimensional current feature extraction, and local joint fault diagnosis; the current detection module has a built-in MCU that periodically collects the raw sampled values of the wind turbine's AC current. .
[0010] MCU local execution processing: 1) Digital filtering: First, perform median filtering to remove impulse noise from the continuous sampled values, and then perform moving average filtering to suppress high-frequency interference, so as to obtain the noise-reduced current sampling sequence.
[0011] 2) Multi-dimensional feature extraction: Based on N consecutive sampling points, the effective value of the current is calculated according to the root mean square formula. ; Zero-point detection was performed on the current signal to extract the current period and frequency characteristics; the peak current was also extracted. Current change rate .
[0012] The effective value of the current is used to determine the shutdown and overload of the wind turbine foundation; the peak current, the rate of change of current, and the frequency characteristics are used to identify belt slippage faults.
[0013] 3) The onboard non-volatile memory stores the fault judgment thresholds of the fan, and the parameters are not lost after power failure.
[0014] The second layer is the current integrated communication module, which serves as an intermediate aggregation and forwarding unit. It is configured with a multi-channel transceiver interface, a main bus transceiver, and a DIP switch. The DIP switch is used to configure the module's communication address, facilitating the addition, deletion, and expansion of wind turbine nodes in the field.
[0015] Workflow: 1) Receive heterogeneous reporting data frames from multiple current detection modules. The messages carry the fan ID, current characteristics, and preliminary fault markers. The message formats reported by different current detection modules are allowed to be different. 2) Heterogeneous messages are uniformly encapsulated into standardized unified data frames; the unified data frame fields include frame header identifier, current integrated communication module address, detection module serial number, fan ID, fan status, current effective value, current threshold, fan fault, and timestamp; 3) The standardized data frames are forwarded to the main control module through the main bus transceiver.
[0016] This module only handles data reception, format normalization, and forwarding; it does not perform fault determination or threshold calculation. This enables fault isolation of a single current detection module, and a single node malfunction will not block data uploads from other wind turbines.
[0017] The third layer: the main control module, which is the main control unit for the poultry house environmental control and is connected to the HMI human-machine interaction touch screen; the main control module periodically broadcasts fan command status frames to all current detection modules, and the fan command status frames carry the target on / off commands for all fans.
[0018] The main control module implements a dual closed-loop threshold management mechanism of manual and adaptive approaches: (1) Manual configuration on touch screen: Operators enter the rated current, zero return threshold, overload threshold and other fault judgment thresholds for each fan on the HMI touch screen; (2) Parameter distribution: The main control module distributes the configured threshold parameters to the corresponding current detection module through the communication bus, and the current detection module stores them locally. (3) Adaptive weighted iterative update: With the adaptive function enabled, the main control module receives the effective value of the steady-state operating current of the wind turbine and constructs a data array of length An; only the steady-state data after the wind turbine starts for a preset time is written into the array; after the array is full, the maximum and minimum values are removed, and the average value Avg is calculated; the weighted iterative formula is used. Update threshold; Set threshold upper limit Threshold lower limit The updated threshold must meet the following requirements. Clear the array and iterate; weight coefficients It can be configured on-site, with an example value of 0.02-0.1; the updated threshold is then reissued to the corresponding current detection module.
[0019] The current detection module receives the wind turbine command status frame broadcast by the main control unit locally, and combines local multi-dimensional current characteristics and local stored thresholds to complete local joint fault determination and generate a preliminary fault mark. The fault mark is forwarded to the main control module through the current integrated communication module along with the message. The main control module parses the fault information and executes pop-up window and audible and visual alarm prompts.
[0020] This invention can identify four types of faults. Fault determination incorporates time windows and multiple sampling for jitter reduction to avoid false alarms caused by instantaneous noise. 1) Current return to zero fault (cover belt completely broken condition): Fan command is on, Interval within time Sampling, number of samplings If the effective value of the current remains below the zero-return threshold, a current zero-return fault is determined. 2) Overload fault: Fan command is on, dynamic duration Inner Spacing Sampling, number of samplings If the effective value of the current continuously exceeds the overload threshold, an overload fault is determined. 3) Belt slippage fault: The fan command is on. Interval within time Sampling, number of samplings A necessary condition is that the effective value of the current remains within the range of 55%-75% of the rated current; if the peak current fluctuates significantly and / or the rate of change of the current deviates significantly, the belt is judged to be loose and slipping; if the belt breaks completely, the current returns to zero fault is classified as a fault. 4) Abnormal fan start-up fault: The fan command is off. Interval within time Sampling, number of samplings If the effective value of the current continues to be greater than the zero-return threshold, it is determined that the fan is abnormally starting.
[0021] Furthermore, the system adopts CAN bus communication; the main control module broadcasts a fan command status frame every 100ms; and the current detection module performs 100 current samplings per second.
[0022] The present invention also provides a distributed poultry house fan status monitoring system for implementing the aforementioned monitoring method; the system includes multiple current detection modules, a current integrated communication module, and a main control module; Multiple current detection modules, each corresponding to a poultry house fan, are used for fan current sampling, digital filtering, multi-dimensional current feature extraction, and local joint fault determination. The current integrated communication module connects to multiple current detection modules to receive heterogeneous reported data frames, encapsulates them into standardized unified data frames, and forwards them. The main control module communicates with the current integrated communication module; the main control module is connected to the human-machine interaction touch screen, which is used to broadcast the fan command status frame, receive the threshold entered by the touch screen, perform weighted iterative adaptive threshold calculation with boundary constraints, receive fault information and trigger alarms.
[0023] Beneficial effects: 1. A three-layer distributed architecture decentralizes fault diagnosis to remote current detection modules; the integrated current communication module achieves unified encapsulation and forwarding of heterogeneous messages; faults in a single current detection module are isolated from each other, without affecting the monitoring of other fans; adding new fans on-site only requires installing current detection modules, and the address can be configured via DIP switches for access without modifying the main control hardware, resulting in strong scalability and reduced maintenance costs. 2. The main control broadcasts the target control command for the fans to each current detection module, which then combines the control command with multi-dimensional current characteristics to perform fault diagnosis locally, reducing the amount of raw sampling data transmitted on the bus and lowering the main control computing power and communication bus load; relying on the joint verification of "command + multi-current characteristics + time debouncing" significantly reduces false alarms and missed alarms caused by noise and power grid disturbances in the poultry house. 3. A dual closed-loop threshold system is implemented, combining manual configuration with weighted iterative adaptive averaging. Adaptive updates utilize extreme value removal and weighted averaging iterations, while upper and lower threshold constraints are set to prevent unlimited threshold drift due to fan aging. The weighting coefficients are configurable, balancing threshold following ability and anti-interference capability, adapting to the mixed deployment of multiple fan specifications in the same poultry house without requiring hardware modifications. 4. The system distinguishes between complete belt breakage (current returning to zero fault) and belt slippage faults. Multiple features, including RMS current, peak current, and current change rate, are used to jointly identify common belt-related faults in poultry house fans, closely reflecting real-world poultry house operation and maintenance conditions. Attached Figure Description
[0024] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a structural block diagram of the distributed poultry house fan status monitoring system described in an embodiment of the present invention; Figure 2 This is a flowchart of the main process of the distributed poultry house fan status monitoring method described in this embodiment of the invention. Detailed Implementation
[0025] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.
[0026] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.
[0027] Example 1 Implementation of 36-way fan poultry house on-site: A large-scale broiler poultry house has 36 ventilation fans with different power ratings. The distributed monitoring scheme of this invention is adopted.
[0028] For hardware deployment, please refer to Figure 1The system is equipped with 3 integrated current communication modules, each supporting up to 12 current detection modules; a total of 36 current detection modules, each connected to one wind turbine; the entire system uses a CAN bus with a baud rate of 250Kbps.
[0029] The static structure of this monitoring system is as follows: each current detection module is arranged near the fan side to collect the corresponding fan current signal; all current detection modules are electrically connected to the corresponding current integrated communication module; the current integrated communication module is connected to the main control module in the poultry house front-end control cabinet via CAN bus; the main control module is externally connected to an HMI human-machine interface touch screen; each current detection module has a built-in MCU and non-volatile memory; the current integrated communication module is equipped with a multi-channel transceiver interface, a main bus transceiver, and DIP switches.
[0030] See the main flowchart Figure 2 The specific work process is as follows: S1. System power-on initialization: The operator connects an external HMI touchscreen to the main control module and enters the rated current, zero-return threshold (example 0.1A), and overload threshold for each fan. The adaptive function is enabled, the weighting coefficient α is set to 0.05, and the upper and lower limits of the threshold are set. 70% of the rated current The rated current is 120%; the parameters are sent to each current detection module via the CAN bus and stored in the onboard non-volatile memory.
[0031] S2. The main control module broadcasts a fan command status frame to all 36 current detection modules at a 100ms cycle. The frame carries all fan start and stop target commands.
[0032] S3. Each current detection module operates independently, with the MCU performing 100 current samples per second; median filtering and moving average filtering are applied to the sampled data for noise reduction; the effective value of the current is calculated, the zero-crossing point is extracted to obtain the frequency, and the peak value is extracted. Current change rate .
[0033] S4. The current detection module locally receives the fan command status frame broadcast by the main control unit, and performs fault determination based on local current characteristics and locally stored thresholds: Example 1: The fan command is on, the fan belt is completely broken, the effective value of the current is continuously lower than 0.1A, and the condition is met for N1 consecutive samplings. The current returns to zero fault is determined; a fault marker is generated and uploaded.
[0034] Example 2: The fan command is on, the belt becomes loose and slips; the effective value of the current stabilizes at 62% of the rated current, while the peak current fluctuates periodically; the belt slippage judgment condition is met, and a belt looseness and slippage fault mark is output.
[0035] Example 3: The fan command is off, but the detected effective current value is continuously greater than 0.1A, which meets the sampling count condition, and the fan is determined to be abnormally turned on.
[0036] Example 4: The fan command is on, but the current continues to be higher than the overload threshold, indicating an overload fault.
[0037] S5. The current detection module sends a reporting frame composed of the fan ID, current RMS value, peak value, rate of change, and fault marker to the current integrated communication module. The current integrated communication module receives multiple heterogeneous messages, encapsulates them into standardized data frames, and forwards them to the main control module.
[0038] S6. The main control module receives data frames, collects the steady-state current RMS value 2 seconds after the fan starts, and writes it into an array of length An=120; after the array is full, the maximum and minimum values are removed, the average value Avg is calculated, and the weighted iterative formula is executed. The result is forcibly constrained within the Tmin to Tmax range; the updated threshold is then sent to the corresponding current detection module.
[0039] S7. The main control module identifies fault markers and executes touch screen pop-up window + audible and visual alarm prompts; the main control module broadcasts fan command status frames downwards at a 100ms cycle, and the current detection module reports status information upwards every 500ms, thereby reducing bus message load, and reports regularly regardless of whether a fault has occurred.
[0040] When it is necessary to add a new wind turbine on site, only a current detection module needs to be installed. The current integrated communication module DIP switch is configured with the corresponding address, and the expansion can be completed without modifying the main control program.
[0041] In this embodiment, the time window is set as follows: T1, T3, T5, and T7 are 200ms; T2, T4, T6, and T8 are 20ms. Thus, N1, N2, N3, and N4 are all equal to 10 samples.
[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for monitoring the status of poultry house fans based on a distributed system, characterized in that, The method is implemented using a three-layer distributed framework, which includes multiple current detection modules, a current integrated communication module, and a main control module, each corresponding to one of the poultry house fans. The method includes the following steps: S1. Each current detection module periodically collects the raw AC current sampling value of the corresponding fan. Digital filtering and noise reduction are performed on the original sampled values; Calculate the effective value of the current based on N consecutive sampling points. The zero-crossing information is extracted to obtain the current period and frequency characteristics, and the current peak value is also extracted. Current change rate ; S2. The current detection module will send a reporting data frame carrying the fan ID, current RMS value, current peak value, current change rate, frequency characteristics, and preliminary fault marking to the current integrated communication module. S3. The current integrated communication module receives heterogeneous reporting data frames from multiple current detection modules, encapsulates the heterogeneous reporting data frames into a standardized unified data frame, and forwards it to the main control module. The unified data frame includes a frame header identifier, current integrated communication module address, detection module serial number, fan ID, fan status, current effective value, current threshold, fan fault, and timestamp field. S4. The main control module periodically broadcasts the fan command status frame to all current detection modules. The fan command status frame carries the target on / off control command for each fan. The current detection module receives the fan command status frame locally, combines it with the multi-dimensional current characteristics obtained locally, performs local joint fault judgment, and outputs a preliminary fault mark. S5. The main control module performs manual-adaptive dual closed-loop threshold management: it receives the rated current of the wind turbine and various fault judgment thresholds entered by the human-machine interaction touch screen, and sends them to the corresponding current detection module to be stored in the onboard non-volatile memory. When the adaptive function is enabled, the main control module dynamically updates the fault judgment threshold based on the effective value data of the current during steady-state operation of the wind turbine using a weighted iterative method, and sets an upper limit for the threshold. Threshold lower limit Apply boundary constraints to the update results; then send the updated threshold to the corresponding current detection module. S6. The initial fault marker output by the current detection module is forwarded to the main control module along with the data frame via the current integrated communication module. The main control module identifies the fault and outputs a fault alarm prompt.
2. The method according to claim 1, characterized in that, The digital filtering and noise reduction in step S1 includes: first, performing median filtering on the continuous sampled values to remove impulse noise, and then performing moving average filtering to suppress high-frequency interference.
3. The method according to claim 1, characterized in that, Step S4, the local joint fault determination, includes current return to zero fault determination: when the wind turbine target command is on, within a set time... Within a time interval Continuously collect and calculate current, and accumulate the data. If the effective value of the current continues to be lower than the preset zero-return threshold, it is determined to be a current zero-return fault.
4. The method according to claim 3, characterized in that, Step S4, the local joint fault determination, includes belt slack and slippage fault determination: when the target command for the fan is "on", within a set time... Within a time interval Continuously collect and calculate current, and accumulate the data. The following conditions must be met: the effective value of the current must remain within 55%-75% of the rated current of the fan; simultaneously, the peak current must also be satisfied. Significant fluctuations and / or current change rate occurred. If a significant deviation occurs, it is determined to be a belt slippage fault; among them, a complete belt breakage fault is classified as a current return to zero fault.
5. The method according to claim 1, characterized in that, The main control module dynamically updates the fault determination threshold through weighted iteration, including the following steps: S51. Collect the effective value of the steady-state operating current after the fan starts for a preset time and store it in an array of length An; S52. After the array is full, remove the maximum and minimum values inside the array and calculate the arithmetic mean Avg of the remaining data. S53. Update the threshold using a weighted iterative formula: In the formula, The current threshold value for the wind turbine before the update. These are the weighting coefficients; the updated threshold. satisfy ; S54. Clear the array, repeat steps S51 to S53, and continuously iterate to update the threshold; weight coefficients. Values can range from 0.02 to 0.
1.
6. The method according to claim 1, characterized in that, Step S4, the local joint fault determination, includes overload fault determination: when the wind turbine target command is on, during the dynamic duration... Inner To collect current at intervals, the cumulative current is... If the effective value of the current continues to exceed the overload threshold, then an overload fault is determined.
7. The method according to claim 1, characterized in that, Step S4, the local joint fault determination, includes the fault determination of abnormal wind turbine start-up: when the target command for the wind turbine is off, a set time is set. Inner To collect current at intervals, the cumulative current is... If the effective value of the current continues to be greater than the zero-return threshold, then the fan is determined to be in abnormal start-up fault.
8. The method according to claim 1, characterized in that, The current integrated communication module is equipped with a DIP switch, which is used to configure the communication module address and to flexibly add, delete, and expand the field wind turbine nodes.
9. The method according to any one of claims 1-8, characterized in that, The system uses CAN bus communication; the main control module broadcasts a fan command status frame to all current detection modules every 100ms; the current detection module completes 100 current samplings per second.
10. A distributed poultry house fan status monitoring system, used to implement the distributed poultry house fan status monitoring method according to any one of claims 1-9, characterized in that, include: Multiple current detection modules, each corresponding to a poultry house fan, are used for fan current sampling, digital filtering, multi-dimensional current feature extraction, and local joint fault determination. The current integrated communication module connects to multiple current detection modules to receive heterogeneous reported data frames, encapsulates them into standardized unified data frames, and forwards them. The main control module communicates with the current integrated communication module; the main control module is connected to the human-machine interaction touch screen, which is used to broadcast the fan command status frame, receive the threshold entered by the touch screen, perform weighted iterative adaptive threshold calculation with boundary constraints, receive fault information and trigger alarms.
