Due to the time-varying, hysteresis and nonlinear characteristics of fertilizer concentration control in the water–fertilizer ratio control system, common control algorithms such as PID and fuzzy PID cannot obtain the expected control effect. In order to accurately control the cotton field water–fertilizer ratio regulation system drip irrigation process of the water–fertilizer ratio that will be controlled within a reasonable range, it is needed to design a bat-optimized variable-domain fuzzy PID water–fertilizer ratio control strategy, through the use of bat algorithm to find out the optimal expansion factor and the best domain of the current conditions, and then according to the changes in working conditions to automatically adjust the fuzzy control of the domain, through the control of the valve openings to change the fertilizer pump back to the amount of water. Realize the fast and precise control of fertilizer concentration in the water–fertilizer ratio control system. Comparative tests were conducted to verify the traditional PID, fuzzy PID, variable domain fuzzy PID and bat-optimized variable-domain fuzzy PID control algorithms. The results show that: if the water–fertilizer ratio is adjusted to 50:1 from the startup, the adjustment time required to reach the target water–fertilizer ratio under the bat-optimized variable-domain fuzzy PID control is 15.29 s, and the maximum overshooting amount is 16.28%, which is a smaller adjustment time and overshooting amount; if the water–fertilizer ratio is adjusted to 40:1 from 50:1, the advantages of bat-optimized variable-domain fuzzy PID are more obvious, with the best balance of response speed, overshooting amount and optimal control effect. In terms of response speed, overshooting amount and regulation time, the optimal balance is achieved, showing the optimal control effect. It is proved that the performance of the water–fertilizer ratio regulation system in cotton field under bat-optimized variable-domain fuzzy PID control designed in this paper can meet the actual production requirements, and these findings can help to develop precise irrigation technology for cotton cultivation under drip irrigation conditions.

The application of water-fertilizer integration and intelligent irrigation systems in cotton production will greatly promote the increase of cotton yield and quality, which has significant application value for cotton production. This paper introduces Internet of Things (IoT) technology in the water-fertilizer integration and intelligent irrigation system and shifts cotton production to an information-centered production mode. In this paper, water-fertilizer integration adopts the first part of the main pipe fertilizer premixing system, which combines venturi and centrifugal pumps to form a fertilizer mixing equipment and uses sensors to collect the environmental information of cotton growth and development and the growth condition of cotton and also designs the fuzzy PID automation control module to realize water-fertilizer integration and smart irrigation. In the application test on June 22, 2022, the air temperature decreased, humidity increased, and light intensity gradually decreased after 15:30 due to the weather turning cloudy. After the rain stopped around 20:00, the sensors detected that the air temperature reached the lowest value of the test practice section, 28°C, which is sensitive to the environment. Compared with the artificial irrigation method, it can increase cotton yield by 66.98% while saving water by 11.59%, and the application found that the EC value of the fertilizer solution in the fertilizer mixing bucket reached a steady state at about 150s, and the fertilizer decision-making model also has a greater superiority compared with manual.