Abstract Effective water management in underground mining requires maintaining safe reservoir levels while minimizing the high energy costs of continuous pumping. Although flexible electricity pricing enables cost-aware operation, traditional threshold-based controllers cannot exploit this flexibility efficiently. This paper presents an industrial case study on efficient mine dewatering using reinforcement-learning-based control synthesized with the Uppaal Stratego framework. A baseline threshold controller is first implemented, followed by a reinforcement-learning controller trained on forecast inflows and day-ahead electricity prices to minimize pumping costs while limiting pump switching. To ensure safety during learning without distorting the optimization objective, we introduce a pre-shield that blocks unsafe transitions. We formally show that this pre-shield is maximally permissive with respect to a monotonicity safety objective. Simulation results demonstrate that the learning-based strategy reduces total energy consumption by up to 40% compared to threshold-based control, while maintaining safe operation in all scenarios.

This paper investigates a new approach to adaptive control for the Precision Time Protocol (PTP), which is one of the most interesting clock synchronization solutions for industrial applications. An adaptive Kalman filter method for PTP control is proposed that jointly considers Wi-Fi performance variations and frequency instability of clock oscillators. We report on first evaluation results as compared to state-of-the-art approaches in the field and highlight future work steps.

The development of wireless communication has indeed promoted cloud-fog automation in the industry. It also introduces new issues of reliability and latency for control systems. This study sought to investigate the impacts of the commonly used industrial wireless network on the control performance parameters using a ball-and-beam (BB) time-critical balancing control system. An internal model control-inspired latency-aware wireless control framework (IMC-LA) is presented and employed in the BB system to handle the time delays and instability-creating elements introduced by wireless communication. A preliminary control assessment of the BB system under two new-generation wireless technologies, Wi-Fi 6 and 5G, was delivered. The correlation between network performance and control performance was analyzed statistically compared to the wired Ethernet condition. Test results show a dramatic decrease in position error after utilizing the proposed latency-aware wireless control framework. This study provides practical insights into the potential impacts of industrial wireless networks on control systems with a workable latency-aware wireless control approach. The methodology presented in this work has the potential to expedite the adoption of wireless communication in time-critical control. Note to Practitioners—Many factory automation processes experience undesirable latencies when transitioning from classic architecture to cloud automation architecture, particularly with the implementation of wireless communication. This paper aims to address the potential instability problem introduced by practical wireless networks and proposes a latency-aware control framework using the concept of internal model control. Meanwhile, this work also provides a way for practical operators to explore the relationship between critical parameters of communication networks and control performance parameters. The purpose of this paper is not to judge which wireless technology is better, instead we only want to demonstrate the effectiveness of the proposed latency-aware control in improving control performance under various wireless scenarios. The proposed framework is verified using the BB system for the commonly used cascaded PID control under two advanced wireless networks, 5G and Wi-Fi 6. It is also applicable to other industrial wireless networks with millisecond-class latency in the other regulatory control cases. The proposed IMC-LA wireless control framework reduces the significant effort and cost associated with constructing and tuning the controller in a real control system.

The main purpose of this work was to further develop and evaluate non-periodic sampling schemes, focusing on their needed communication and the effect they have on the control performance. The paper focuses on evaluating the schemes in experiments on some real industrial processes at Iggesund Paperboard. The industrial evaluation was done using fast wired communication into an ABB AC800M controller with a pre-filtering block mimicking non-periodic communication. The tests show communication reductions of 90 to 99 % compared to the current default sampling intervals in the industry, without significant loss of performance. However, the results further indicate that often carefully selected slower periodic sampling may reduce almost as much communication. In the end the choice of method for communication reduction will be determined by the implementation effort of non-periodic sampling versus the commissioning effort of slow periodic sampling.

The WirelessHART standard uses TDMA and channel hopping to control access to the network and to coordinate communication between network devices, in order to enhance reliability and to improve the throughput of the network. A problem in utilizing multiple channels is that current devices are usually equipped with a single transceiver. Thus, a node can only transmit or receive on one channel at a time. Moreover, contrary to today's wired control systems, if a single access point is used the communication becomes the bottle neck of the control system. Therefore this paper presents how one may schedule the WirelessHART communication using two access points. Furthermore the paper describes a scheduling algorithm managing a multihop multi-channel networked control system based on the WirelessHART standard. A simulation example of a multihop multi-channel network is also shown, using the fixed packet lost utility of the Matlab/Simulink-based tool TrueTime.

This work describes simulation of wireless control using the WirelessHart standard. Specifically issues concerning clock drift between controllers and wireless network are addressed. The simulations have been done using an extension ofthe Simulink package TrueTime.

There are two approaches to the identification of noncausal autoregressive systems in two dimensions differing in the assumed noise model. For both approaches, the maximum likelihood estimator formulated in the frequency domain is presented. The Fisher information matrix is evaluated and found to be the sum of a block-Toeplitz and a block-Hankel matrix. The variance of the parameters, however, cannot be used for comparison of the two approaches, so the variance in the frequency domain is evaluated, assuming that the true system in each case can be described by a model of that type, possibly high-order. In particular, the variance of the spectrum estimate is derived. If the number of parameters tends to infinity, it is shown that the two approaches give the same spectrum estimate variance. The question of which set of true spectra can be described by the respective approaches is discussed.>

The concept of tree classifiers is combined with the popular neural net structure. Instead of having one large neural net to capture all the regions in the feature space, the authors suggest the compromise of using small single-output nets at each tree node. This hybrid classifier is referred to as a neural tree. The performance of this classifier is evaluated on real data from a problem in speech recognition. When verified on this particular problem, it turns out that the classifier concept drastically reduces the computational complexity compared with conventional multilevel neural nets. It is also noted that these data make it possible to grow trees online from a continuous data stream.>

A study is made of parametric estimation of 2-D transfer functions using the least-squares method which is the dominant method for causal systems. The analysis concentrates on the frequency-domain accuracy of the estimated models. There exist results for the accuracy of the parameter estimates. The parameters are asymptotically Gaussian distributed. The variance of the polynomial in the frequency domain can be expressed using these results for the parameters. This, however, gives no insight into the dependence on the true transfer function. An illuminating result is obtained if the model order tends to infinity. The limiting results show good correspondence with Monte-Carlo simulations for small data sets obtained with low model orders.>